I often state that WetWare is a key to software tool/technique productivity or "worth" analysis and that software engineering is not really about computer hardware or some kind of "pure" science. However, WetWare is still a fuzzy science such that objective analysis is difficult. 

Better models are needed to make judgements on it. However, if the model is good enough to "run" in realistic situations, then essentially we've created AI that closely models the human mind, and which point we will all be out of jobs soon, making the findings moot.

Maybe the first couple of years it will be a mere "lab curiosity", but as soon as such AI is demonstrated to actually work, investment money will pour in, making it a mass-production technology in only a few years. The cost per brain equivalent would only need to drop to around $80k a year, USD, to be competitive with human software-related employees, including administrative overhead associated with human employees.  (Investment money poured into AI in part of the 80's even though the results were less than half-baked. Imagine investor reactions to a fully functioning one.)

--top