in the current build of the game, one way to combat the rising unrest percentage from the number of cities is to build some extra fortresses and get them to level 4 ASAP and pick the "prison" upgrade for -10% global unrest. this mechanic works, but its rather crude in my opinion and i think it could be improved by making the process more gradual.

the way I see it, getting a city to level 4 is strictly a late game thing, and frankly, at that point most of my games are already at the stage where it doesn't really matter any more- i' m either winning or losing that game, a few % more research/production across the board won't change that outcome much.

i think using fortresses as centers of unrest control is a good idea, as it gives them more purpose. my suggestion is to move the unrest reduction from the level 4 upgrade to other buildings that you can build up earlier. the defensive structure line (hedge walls, castles) seems like a good candidate for this, as it's also thematically somewhat related (walls as a symbol of stability - make the population of the kingdom feel secure knowing that their borders are secured by mighty fortresses etc; or the other way round - intimidating walls as a symbol for the futility of rebellion that help oppress rebellious tendencies in the heads of your empire's citizens)

an implementation could be something like - hedge walls reduce number of city unrest by 2%, castles add another 4% and one of the level 4 upgrades could be another defense structure that adds another layer of defense (more /better defenders) and also another boost to unrest management (another 4% maybe). this could also be a level 3 building to add a difficult decision - do i want the free charge/endurance or do I use the fortress to handle my unrest situation better. (free impulsive should be removed or delayed to level 4 anyway - way too powerful imo)

so you basically end up in the same situation as now, but you get part of the bonus along the way and not all of it at the end. the early hedge wall/wooden wall basically removes the unrest caused by the fortress itself, and the castle removes another 2 cities from the equation; the level 3 or 4 building ("stronghold" maybe? ) then removes another 2 cities from the calculation.

I'm aware that there is also the 30% reduction from the level 5 upgrade (onyx throne i think?), but i don't think that's much of a factor. tbh, i only ever had one level 5 fortress in all of my ~300 hours of FE, and that was in a game that was long decided before i finished it. in fact - i delayed victory in that particular game with the purpose to get my core cities to level 5 to see what upgrades i'd get. i don't think level 5 cities are common enough to factor into that proposed mechanic.

EDIT: just had another idea for the level 3 or 4 replacement upgrade: it could be something like an "armory" that boosts city defender units (milita/militia archers) by 1 level empire wide (similar to the empire wide boosts for high level towns and conclaves). stack multiple fortresses with armories to get some decent militia units (with extra HP + accuracy from the extra levels)