Very little, I think. Maybe sometime in the future some great technology will arise that may help us make sense of it, but I'm skeptical.
My take on this topic - which tends to be controversial - is that aside from hints and suggestions, looking at the brain tells us very little about higher cognitive faculties. It's not nothing, obviously, but little in terms of what we would like to know, such as the question you are asking.
What's curious here, about this activation pattern, is that (I don't think it's in this book, but in another essay whose name I've forgotten) similar sounding noise doesn't activate it. For instance, if I say:
Under space roaring goes doesn't anywhere nothing.
Here each individual word makes sense, but the sentence is gibberish.
On the other hand, if I quote Chomsky's famous:
Colourless green ideas sleep furiously.
The sentence makes syntactic sense but doesn't mean anything. I'm blanking on the study, but if I find it, I'll post it here.
When they do tests with subjects, they show them ordinary languages that they don't know. If it's a human language, the brain activates. But if they produce sentences that breaks these rules, the subjects don't register it as a language.
This of course leads to even deeper questions, such as, why don't we register every sound as something significant and meaningful and say, don't confuse others sounds with language? There must be an innate property we have, that accounts for this.
So other than a general comment about, human language being an extremely sophisticated, unique to humans' phenomena, I can't really answer the question.