I do not think there is much neurological understanding about ADHD at all from current fMRI research, there are far too many quality and reliability issues here, not just on the fMRI end or limited amount of data overall, but in the measurement and diagnosis of ADHD itself (i.e. ADHD subtypes, and of course ADHD is a complicated diagnosis with many components manifesting to different degrees in different individuals, which makes it very hard to cleanly link to messy fMRI signals).

Or, as I have commented elsewhere here, the idea that statements like "fMRI shows decreased activity" are ever valid is just fundamentally suspect (lower BOLD response could mean less inhibition or less excitation, and this is a rather crucial difference that fMRI simply can't distinguish). EDIT: Or to be more precise: it may well be that fMRI research suggests less metabolic activity in certain regions, but this could mean the region is actually firing more than normal, less than normal, is more efficient than normal, etc., and interpreting anything about what is functioning differently in ADHD, given this uncertainty, is what is going to be suspect.

Your analogy is largely correct IMO.