I did something similar as part of learning English. For me the best author for this purpose was Stephen King (not that I spent much time searching, just random trial and error; I don't even like horror stories). After mastering Stephen King I moved on to J. R. R. Tolkien. For the first fifty pages or so of The Lord of The Rings I had to look up words pretty much one for every other line of text.
It was amazing to see how different authors, although both very good and well-known, use language so very differently, and use quite different sets of words and of very different sizes. It's not like Stephen King writes children's books, or that his stories are any more simple than LOTR. It's just like with code: You have people using a small(er) subset of a language but express similarly complex ideas (algorithms), while others enjoy using less commonly used language features. Both methods can get the job done beautifully in the right hands. One has the advantage of being more (easily) accessible (and to a larger amount of readers) though.
By the way, my next level in "word acquisition" came through reading The Economist and a few other magazines of similarly exalted vocabulary. Again I had to look up at least ten words per article when I started.
Which reminds me: Is there a service that manages to categorize books by looking at the set of words they contain (and possibly sentence structure/grammar)? Might be quite useful as a niche service for foreigners learning a language. The obvious and simple answer is to sort by intended audience (e.g. books for children and young people vs. sciency books), but as my Stephen King vs Tolkien example shows, the overall level does not have to be different at all just because one uses a smaller set of words.