Math is like chess. You really improve by playing long games or solving tough problems. Tests are often about speed and not getting things confused. So you might want to step up a notch with challenging problems. It will make standard problems look much easier and less confusing. In my opinion you should treat them like bullet chess that is solving them as fast as possible with as few mistakes as possible. The only path is concentration on tough problems. Just like in chess: Playing bullet games don’t improve your strength at all; you feel they dumb you down. The same is with math tests where speed counts a lot. It feels like they dumb you down and only speed is required! The deeper you go, the more challenging things you do, the more confidence you’ll have. Just don’t focus on simple (test-like) problems. It’s just like focusing on bullet chess. You may play 10,000 games without much improvement. That might create anger, anxiety, and a lot of negativity. You work hard on speed (or on standard problems) and get no results. That’s terrible anxiety and it’s very discouraging. I mean doing a lot of problems without the necessary results on tests. Of course you will need to work on speed too, not too much though, but it is necessary. It is just like in bullet chess when you have to make a decision fast, although the move requires a lot more.
So, only real strength and superiority can boost your confidence and diminish anxiety. Playing fast won’t cut it (=getting only standard and confusing questions right in time trouble). Conceptual understanding, challenging questions and super subtleties will get you the results you want. Tackling anxiety won’t cut it, although it might help. I think you know that one textbook and one solution manual might be just the tip of the iceberg. Anyway I'm gonna emphasize here that sometimes you need a more advanced book along with a simpler book to get better results. Training for speed and tests are secondary, although it is important too. Treat tests like funny bullet chess games with haphazard results, and treat challenging problems and tough conceptual things and subtleties like real knowledge and strength.
Bottom line is you always want a serious book (+a simple one) and serious problems to tackle. A serious, challenging solution manual is a must too. If it is already tough as it is, well, the science is getting tougher and tougher all the time. More and more time is needed to cover all the bases.
Just my opinion.
A couple of funny quotes (New edit: probably should not have included these quotes as they draw attention to some humor and they are beside the point. They were not meant to create some negativity here or present any truth here, though they main contain something of the kind)
“A mathematical theory is not to be considered complete until you have made it so clear that you can explain it to the first man whom you meet on the street”
“Never listen to anything your math teacher tells you” (said by an anonymous math teacher)”
Really hope that my advice won't be useless to you.