If you prefer to use LaTeX, use LaTeX. My preference is to write content in Markdown---which is almost as technology-agnostic and ubiquitous as plain text itself---then figure out the typesetting independently.
For what it's worth, even though the blog discusses Markdown and ConTeXt, much of it applies to Markdown and LaTeX.
By the time you reach the R section, there are only three scripts: build-template, which is reused by the other scripts, the build script, and intarsia. You could also use R Studio, as well, but then you wouldn't have the ease of interweaving interpolated variables from YAML files.
If you find LaTeX's packages and syntax to be simpler than writing (or downloading) a few bash scripts, then go for it. I certainly cannot tell anyone what they'll find simpler to use. For me, Markdown is simpler than LaTeX; I find I can focus on content far easier without having to write or look at macros. À chacun son goût!
I'm really pleased to see you're using Pander[1] in your tool chain. I think this is one of the most useful literate programming tools in the R ecosystem. Although, having read Part 6 (briefly), it looks like you might only be using it for tables and plots. I think Pander's literate programming functionality, in conjunction with Brew, is worth further exploration[2].
Also, if you're using R-base plots, there's an excellent tutorial[3] and theme library[4] which can help make those plots look really excellent.
I didn't find LaTeX simpler for creating a book with code comments. It's a whole load of stuff to learn, akin to a "Just Learn Kubernetes" level of brain load, which seems OTT when you just want to experiment and create a book.
Could you summarize "why not just use LaTeX?"