It's not like he
hasn't done this before.
That's not quite right. Modern Mandarin
Riben underwent its own sound changes. 日本 was originally
hinomoto (sort of... see below), then later came to be read with its
on-yomi. This has happened to other common words, where Japanese people see two kanji together, figure they should be read with their
on-yomi, and we get two ways to read a word. If you use Japanese dictionaries enough you'll see what I'm talking about. You'll also see the historical kana spelling that would give a better idea of how they used to be pronounced, like かは for 川 (but old ハ行 actually had a "p" initial, so it was "kapa"). A better example would be 平等, which shows its old spelling of びやうどう (it change via palatalization and the au -> ô change). At any rate, using modern words in languages to show connections is a bad idea, because they aren't now what they were then (using modern words in one language with ancient words in another is also a bad idea).
Particularly regarding 本, and kanji in general, their present forms have changed greatly from the original forms. If you want to know the origin of a character, you have to go to the source, which is oracle bone inscriptions in most cases, where 日 looked more like a circle with a dot in it, 馬 looked like a horse, 川 a river, and so on, but then they changed their shape as they moved away from pictographs and more towards a writing system (the writing materials also had an influence). This is why you can know Chinese or Japanese well, and not be able to make out what a character is in oracle bone script (different from grass strokes, which presents its own difficulties). It gets more involved when you start looking at how they were pronounced 5000 years ago.