Good stuff, looks fantastic. The only major thing I'd say is really missing is the human aspect of regions/borders. Assuming the regions in-game are areas that are discovered, named, categorized and outlined by the polities who discover them, it seems strange that they can map out an entire region far beyond not only the borders of their polity, but can even map out regions generally uninhabited/uninhabitable. I point this out not because its a huge flaw, but rather because it highlights some possibilities.
For one, the region exploration system is not dynamic (as far as I know). Once a polity happens to touch a cell that is not contained within an already defined region, the algorithm outlines a new region, names it according to the language of the discovering polity (I think) and thats that. That region will never shift, never change, the name remains the same even as the people who inhabit it come and go and even as their understanding of the region in regards to the greater world increases, the region will always be the same. This does not reflect how human beings regard regions/territories in the real world.
We define our regions according to a myriad of factors other than bio-geographic properties, for example we define regions by the culture of the people who live there in comparison to those around them, to the history of that region and for political reasons (often territorial claims or plain logistics (I.E regions that are absurdly large are difficult to manage at a certain technological ability)). All of these factors change through time, they are by nature dynamic, irrational, but also useful and interesting.
I'm not saying you should create an advanced sociological model of human behavior before you alter how regions behave, but rather a system should be put into place to account for the dynamic and changing nature of human borders put upon the world, and also the limitations of early human exploration.
Regarding regions, when a region is first explored and settled by a polity, it should still of course be defined by bio-geography, for this is what said polity will be worried about and therefore will be how they categorize the region. Then as the next phase of civilization begins, and small yet numerous polities start to compete for land, the focus of the people is now who lives where, for how long and what polity to they belong to. As such the regions borders begin to shift (and in accordance with adjacent regions) to match cultural borders. This reflects the changing nature of peoples understanding of the regions in which they live. This change should be more gradual (not tens of thousands of years, but not a few months either). As tribes migrate to and from this region it may split, as the more ferocious tribes carve their corner of the region, over time the neighbors should consider it to be its own region, and should that ferocious tribe settle down and cooperate they and the land they live on once again reincorporate into the region.
This dynamic behavior can be attributed to sub-regions with regions being a little more grounded in geography, however I do think sub-regions should be able to their own regions which can begin incorporating other sub-regions depending on certain factors.
Whatever the specifics are, regions and sub-regions shouldn't be a static backdrop as they reflect the understanding of the people who live within them, and their should be a basic exploration mechanic to reflect the time and organization needed to map out the world.