Oftentimes chessplayers will feel that they hit a wall in their chess development, or simply want to improve faster. So a question that many like to ask to grandmasters is, "What do you do to get better at chess?"
There is no secret ingredient for chess improvement, but to satisfy the curiosity of anyone who is interested, I will provide some description of my chess schedule. What I will not provide is a numerical value on any of the activities. I don't like to be asked how many hours per day I spend on chess training, because it varies greatly from day to day, and I keep no record, so I can't answer within a reasonable margin of error.
I never force myself hold to any strictly defined training regimen like 'solve tactics for 1.7 hours per day', since that probably wouldn't be very fun. The two things I do the most are analyzing with chess engines and playing internet chess. When I am analyzing a position with the computer, it is usually an opening. Also I sometimes analyze my own games (with or without a computer), and random positions from games I'm looking at that catch my eye.
So openings are the phase of the game that I study the most. I used to rely heavily on books, but now I learn more by using ChessBase and computer analysis. I easily get lost spending 3 hours to find the path to advantage in some position which can arise at move 19. Opening analysis consumes more time than anything besides internet blitz (or playing tournament games). Probably this is why I'm kind of bad at the aspects of chess other than the opening. I do the best in the opening and early middlegame (though opening disasters still happen). When my opponents get past this point without getting annihilated, things often start going downhill for me. A common theme is that I will have a serious advantage early on, and then I will gradually run my position into the ground until it is a draw or a loss.
Sometimes I do tactics problems. Most recently I am doing the ones on chess.com. These are mostly ones that can be solved in just a few seconds, so it is not really intense calculation training. But I really like chess.com tactics when I am falling asleep in a class, because they help me to stay awake. I do own a few books with positions that require longer calculation.
I spend little time studying endgames. In 2008 I worked through Dvoretsky's endgame manual, but I have not done too much since then. I find it unrewarding to study theoretical endgames, because I rarely get any of these positions in my tourament games, except for the most simple ones like king and pawn vs. king.
The thing I do most consistently is to read the daily email newsletter Chess Today. Usually it contains annotations of a game or some game fragments or endgames. It's a nice way to squeeze in some chess any time when you have a few minutes to spare.
Chess books: I have some books about various things, with the most common topic being the opening of course. I'm not going to discuss specific books here or give any recommendations; it is too broad of a topic. Certainly reading chess books is not a bad idea.
If I have no specific ideas for necessary components of chess training, are there any dangers to avoid? There is a variety of theories about things that are bad for your chess. For instance, some people have said that playing blitz makes you worse at chess. I think this is nonsensical. Being an internet blitz addict, I like to think that it helps me develop a feel for how to play the middlegame. Though I must admit that it does little for endgames, where the main skill is flagging. However, I do try to avoid playing a lot of 1-minute games. Even if they don't make one's chess abilities deteriorate, it can hardly help them, and to me it feels like a waste of time.
Another idea which seems like common sense is that you should avoid playing games like bughouse which, could interfere with your thinking in normal chess. For example, you might start a sacrificial attack, not remembering until too late that you are not allowed to deliver the coup de grâce by dropping a pawn on g7. I say don't be too worried. In fact, I was playing a ton of chess variants (bughouse, losing chess, atomic, crazyhouse, etc.) during the time when I was gaining rating points most rapidly. So personally I have found that it is not harmful to play these.
My advice is that you should have fun playing chess by studying or playing whatever you find enjoyable. My other recommendation is that you don't take the above remarks too seriously. Really, I don't know anything about chess improvement