Yes,you are right about Yusupov-finding mistakes in his books is very rare (In fact so far that was the only one,and the computer gets it wrong too until 2 or 3 moves into the variation,meaning that he takes writing his books very seriously,and also that he checks it with a computer!).

Having said that,some books (by GMs/IMs) do contain errors that are not even subtle :)  

I will give a few examples soon if you want,when i will have enough interesting positions to show!

I'm surprised GM Yusupov missed that -- though the idea is subtle, it's not like it is completely alien to me, so you'd think it would be downright obvious to him that you have to look at what happens in the position after the queen moves to a5 to make sure there are no last second ideas. I suppose he either A) figured that the queen being trapped was too unrealistic, so unrealistic it wasn't even worth mentioning in his notes; or B), he was simply a little lazy there.

Or maybe a third possibility: He computer checked it and perhaps the computer also thought it was just good for white :) Anyway, nice find, if it works -- must be fun to "outsmart" a grandmaster for a millisecond ;) Actually, now that I think of it, I guess he assumed that if ...Bc8 didn't work, due to giving up a4, then the trap wouldn't work. Still, even though ...Bc8 doesn't work, the queen is still trapped and out of play with the bishop on d7, even though it will be a while before it finally gets attacked.

I remember once where I was reading a book and GM Igor Stohl said "and black has dangerous attacking chances for the piece," but I thought he was being too presumptuous (A weak king and active pieces doesn't automatically equal mate), so I looked into it more with the help of a computer, played it out quite a bit, and I came the conclusion that black's attack was basically BS, and that he was in trouble.

Of course, finding such mistakes is like seeing a top player make a huge blunder -- it's psychologically encouraging because it shows you they are human, but at the same time we know that such mistakes are extremely rare, and we make them way more than they do.