You won't always have an optimal solution but that's okay. The most important is to try and use the thing you're learning in some real way or with practice.
In my experience, that's a necessary first step to learning. I need to get my hands dirty, get a feel for what I'm working with. An experience is worth a thousand pictures, which are worth a thousand words - you can't gain that basic understanding and instinct by reading, only by having all the sensory inputs of doing it.
Then it's time to read. Now you must find an expert to guide you. First, you'll have too many blind spots - you can't possibly find all that's current, you can't find the best sources efficiently, and much won't be in books yet. And without expertise yourself, you can't distinguish the worn-out theories from the evergreen standards from the unproven innovations; the promising from the unlikely from the absurd; you won't know the consensus from the fringe; the guy advocating their personal theory - maybe even a credible one - from the balanced survey of established ideas. You won't recognize when you're reading just a side of a well-known debate.