This was not at all what I thought it was about from the title. The TL;DR: You can read more than one book at a time on various subjects.
What I thought it was going to be about: When reading non-fiction (in particular) you don't have to read the book straight through. Since this is HN, take a book on your hot new language of choice that you're going to submit 5000 posts about here. If you're already a programmer, the first 4 chapters or so are likely pretty useless to you other than for a quick skim: Here's how you declare variable, functions, classes, traits, whatevers. Here's the syntax for various expressions and/or statements. Takes a day, at most and you may not read it straight through, just skip around looking at the examples, look at the text if it doesn't click, type up a few examples/test cases and see if you grok it. Then the next 6-50 chapters detail specific aspects of the language, maybe a standard library or how a particular tool or feature is used. You're already an experienced programmer, you grok their OO model (if they have one), but maybe it has a somewhat novel concurrency model. Jump to that chapter. Go backwards in the book if something in it doesn't make sense to the relevant sections.
This can work with most non-fiction depending on your level of familiarity with the base subject or the style it's written in. History books sometimes have a narrative structure that makes this harder. A math book may be written with a different enough notation that linear progress is more critical so you can understand the author. But even then, I was reading an astrophysics book earlier this year and I did exactly like I described above for programming except with more attention to the first 2 chapters (calculus and diff eq were ages ago for me) followed by skipping to chapters that discussed things I needed to know for work.