Another issue is that EEG produces massive amounts data over time (since it can detect relatively quick changes). Therefore just analyzing the reams of data becomes a big issue.
I could see combining dry, wireless EEGs (expensive) with a recording camera and some basic computer analysis to reduce the amount of manual human analysis needed. Once could have some brave soul wear the EEG cap and camera through their day (or for some study-specific interactions), and then make a first-pass analysis of the EEG data for "interesting" brain states, and then use the recorded camera data to determine what the person was doing at the time.
Then one could have subjects take various statistical measures... or group them and then try to come up with tests (instruments) like MBTI or Big Five measures that would align with the groups.
On a practical note, EEG is still fairly expensive for individuals: around $6000 US for decent 20 sensor model, and at least $20,000 US for a dry wireless model (some are rentable for around $4,000 US a month, though). Still, over time those prices are going to come down, so it seems inevitable that where will eventually be commercially possible to obtaining EEG data of oneself.
F3 and F4 tend to be a lesser used region by most people (other than TPs), unless they have had training or real-world experience that require the use of those regions. Even Nardi sheepishly admitted that as an INTJ, he probably didn't really use those regions much (which one would think would be critical for logical reasoning).
ISTPs tend to favor T3 (integration of visual-knesthetic data) and P4 (that gambling/economist region that helps weight between multiple alternatives each with their own pros and cons). This makes sense, given the more hands-on, risk-taking behavior of STPs.
One additional difference is humor. Use of F4 (precise definition and categorization) is associated with humor based on surprise and unexpected outcome. F4 humor is associated with incongruity, like puns (whatever one's opinion of them) and witticisms (like Wilde's "A little sincerity is a dangerous thing, and a great deal of it is absolutely fatal").
ISTPs show less activation of F4, and tend towards a humor that graviates more towards breaking social conventions (TPs in general tend not to show a lot of activity in T5, which attends to social feedback).
Of course, sense of humor varies amongst people of the same type, but Nardi did spend some time on the F4/incongruity humor correlation.