Well to be honest, I'm not that far along into probably some of those same books myself, and I still get derailed by random curiosities and end up on Wikipedia

Well to be honest, I'm not that far along into probably some of those same books myself, and I still get derailed by random curiosities and end up on Wikipedia or ordering a new book. The latest was a tangent into filter theory and digital signal processing, for a real-time audio project. It's hard to anticipate everything I should be able to do but I have some idea of the general order of things to study personally. If I needed to go in depth about Linear Algebra (and I definitely do), first I'd recall whatever I know about vectors and vector spaces, transformations and matrices and systems of equations, and start gathering articles and seeing if I'm missing any critical linked topics. After I get confident enough, I'd probably skip right to some textbook topic that caught my attention, say normed vector spaces, and try to make do with backward-skipping as a last resort. Of course, that's just an artificial way of making the problem harder, basically, so that it's more interesting than safer but repetitive sets of exercises. The best way to be interested, by far, is to actually want to solve some problem. For now, anyway, I heartily recommend looking for problems that these disciplines can solve, and choosing one to solve yourself.