Aren't I supposed to get better at this?

Imo, Nimzovich is not an author for beginners.

He had great chess insight, but often others state his ideas better than he does. best get him filtered through other people's eyes.

He does not have enough diagrams in his books also.

Best to study tactical players first. Learn how to play the open game. Then start to try the positional approach of types like nimzo.

Reti did not seem to write a lot of books, but his "Masters of the Chess Board" is very instructive.

CJS Purdy's book is the best for beginners I ever read.

Of course, learn basic King and pawn end games first.

I alo found that mastering the rare Knight and Bishop and King  vs King ending is a great way of learning how pieces cooperate. And if it ever occurs, in about one in 5000 games, you will know how to do it.

If you want to really use your Nimzovich book to advantage, master one part. That deals with the 7th and 8th ranks and use of the rooks. That for me is the most useful part of the book for beginners.

He really is of little help in the openings for beginners, in spite of him having some pretty good opening theory. He just does not explain it well. 

The best lessons are those we learned from our mistakes. When a game is lost, instead of asking for a rematch. analyze the game, the tactics missed, the endgame play and lastly the opening blunders. And always set time to study tactics and games of grandmasters.

It's not really a question of the rating, but setting the tactics below your current level that's the advantage with going unrated. If you are in rating mode and you keep on getting them correct then you get harder tactics and that's not necessarily the best way to learn basic tacticial patterns. It has been suggested by Heisman and many others, that it is better to do lower rated tactics over and over, as opposed to increasingly difficult tactics. The goal is not to 'solve' the tactic, but to burn the pattern into your head so that you can instantly spot the theme when you come across it on the chess board.

True, you may want to look at a few harder tactics now and again, for that you can pop the rating back on or up the tactic rating level and play unrated. Perhaps the most important thing though is not to ever guess the tactic, but to try and see it though to the end.

I've been training with Chess Puzzles by Yan Levesque (android). Managed to get up to a 1940 rating after weeks and weeks and weeks of mucking through the 1500s. I don't know. I'm back to 1730 or something. Studying tactics, realize the problems are intricate, and the positions are many. It's not like you're learning one simple task that you can never forget. Tactics involve such a wide array of specific skill sets, and of course pattern recognition is a big help.

One thing I've learned so far:

1.) Some problems, you see the answer right away during your search of candidate moves, but you actually reject the answer because it doesn't look playable. The consequences initially appear no-good for you, but in fact that's the answer. Sometimes the answer is a move that you think isn't the answer.

It's not really a conundrum. Sometimes tactics training is all about opening your eyes so that one sees that an "unplayable" move is an option.

It can also be useful on defense. I once played a game while making full notes on each move. I was embarrassed to see afterwards that I had specifically disconsidered two crucial moves because I thought my opponent "couldn't" play them. 

Don't make the move if you are not sure it is the right solution. Even if you have to sit 30 minutes in front of a single problem, it is better than do each one in 10 seconds and miss more than half.

I don't think it is good advice to sit in front of a puzzle for 30 minutes struggling, particularlly at beginner level if the idea is to learn patterns. I would spend no more than 5 minutes then look at the solution, then try to learn it. Even better, return to the same problem the next day and recap and so on and repeat.

Well I spent 40 minutes in front of a problem yesterday, a very difficult one, and I solved it. I think that single problem improved my skill more than if I made 40 problems 1 minute each. But it's up to each one to do what they think is better.

I have been using the Chess Tutor for about 6 days now. I get some wrong, most right. I find the Tutor to be decent, but not perfect (sometimes it takes away points when it shouldn't, sometimes it isn't as clear as to what it wants when it should be), but I have noticed that it has improved my Tactics score.

At times, especially being new, I wish I could turn the timer off on Tactics. At this point I need practice more than scoring (I use the iPad version), so I don't fret so much over the score as getting the solution. Unfortunately, my scores suffer at this point in the game (no pun intended).

I would suggest the same, work on getting the problem solved, ignore the scoring. Today I solved my first daily puzzle that wasn't, what I consider, an easy puzzle. I credit the Tutor and the Tactics for improving my skills, 6 days ago I would not have been able to solve it.

Yes- keep trying=the game takes to learn

I offer the following from THIS mentor lesson:

"When a player looks at a position to see what his opponent can do, he often internalizes a checklist where he says, "After I make that move, he can't do that, he can't do that, and he can't do that." Oddly, the other player is also aware of that checklist and, quite often, he agrees! Thus, a silent consensus is formed with both players agreeing on what can and cannot be done. But, what if you refuse to go along with the program? What if you try as hard as you can to make the impossible possible?"

I guess conundrum does means confusing, and hard, so I'll concede the point. But I stand by the idea that learning sometimes means finding a way to make that great move despite your opponent having it defended, especially when your opponent thinks it's well-defended. 

P.S.: That Eli Wallach glare from The Good, the Bad and the Ugly is not a conundrum either. But, it sure is creepy. 

It's Steven Segal... even creepier.

Maybe a more appropriate image of wallach:

Spend more time examining the theory and resolution of T/T problems before attempting them?

Little point in 'rattling-through' them without understanding those you fail.