I would have to dig into the research you posted....but being in the Army city combat is considered the most lethal to this day.
Also fighting in Stalingrad, Berlin, Konigsberg, Breslau, Budapest....would all seem to disprove I think whatever the think tanks did.
Even Kiev in 1941 was a much bloodier affair than the germans thought it was going to be.
The key thing with city fighting is do the defenders have enough men to defend it...which would mean a weighted calculation added to battles I guess..1 division prolly wouldnt have the manpower to hold a large city hex alone this weak spots and gaps can be exploited...2-3 divisions and the other side could only win with frontal assaults in dense terrain.
Just the realistic thing about city fights. Even in modern war city fights are brutal costly affairs..look at Mosul both US vs militias and the more recent Iraqi army vs Isis....even with all the modern warfare gear the battle took a long time and was costly in losses....esp when you factor the tools of modern war that gave a massive edge to the attacking forces.
Well, armies sometimes cling to outdated concepts for a long time. So certainly conventional wisdom would have to be reexamined. But there are no funds for that.
IMHO any kind of a battle tied to a geographical position that cannot be enveloped, with both sides being able to throw more bodies into the furnace of war, can turn out to be lengthy and bloody affair.
Mosul took 9 months and 1500 KIA for the Iraqis (that's 5 per day). It takes time, true (especially if you're trying to avoid losses), but doesn't have to be bloody. Can't compare with battles like Budapest (much shorter but immense casualties). It's really difficult to affect the number of losses produced by WitE combat engine, but it is much easier to affect battle outcome. So at least it could be made so that taking cities (especially not isolated ones) would take several turns instead of one. I'm thinking about borrowing double dense terrain and much higher defensive bonuses for urban hexes from WitW.