Outdoor Survival Map ACKS Style

So, I’ve been reading the campaign design chapter and thinking about the Outdoor Survival Map, and I want to run some numbers to see if they seem feasible.
First off, assumptions: on the original game board there are around 25 catch basins, and in OD&D it says you should put one stronghold for each catch basin, so I went ahead and did that. It’s mostly castles and churches with some a few dwarven vaults, two elven fastness, a bard hall, a few sanctums, and so on.
The concept I’m going for is a mostly untamed land where settlements are far between, but the land is ruled by a bunch of warlords who range from malevolent (even chaotic) to benign and everything in between. There are lots of ruins from the previous inhabitants (an entire culture died off) for the PCs to explore, and they will have to deal with each lord (petty and noble) as they adventure across this land.
The map has a length of 34 hexes and a width of 42 hexes, yielding 1,426 hexes of map (or 45,696 square miles). Since I decided that the dispersed settlement pattern is due to several equal size domains, and the rest being wilderness, I decided that a population density of 25 people per square mile seemed favorable.
I also decided that each stronghold would be a march with about 8 hexes of claimed territory, and thus that makes about 200 hexes of settled land. Combine that with the data above, that gives us a population of 160,000 people in this area (or 32,000 families).
Using the dispersed settlement pattern for urban demographics, as well as the agrarian pastoral realm pattern, that gives us an urban population for the realm of 1,250 to 2,249 families, with the largest settlement being a large village with 250 to 449 families.
So, question 1, how does this all look, and based on your knowledge of the outdoor survival map, does this all sound right? I based 25 people per square mile as a lowball number less than medieval england (which is 40 people per square mile), I originally had thought of 20, but I liked 25.
Question the 2nd, can I use the realm population to determine the largest urban settlement, or does it have to be by the domain population? If I used the realm population, does that mean I just arbitrarily assign which domains get the urban settlements based on geographic favorability?
Question the 3rd, am I right in understanding that the maximum level of the leaders of these domains is based on their realm population? So you could have NPCs less than 9th level ruling these domains?
Thanks for taking the time to read and answer my questions! I’m really enjoying the game!

Question 1: It looks good and sounds all right!
Question 2: You can absolutely use the Realm Population to determine the largest settlement. The Village, Town, and Cities Placement are just guidelines and you can always over-rule them as necessary. You would assign which domains get the urban settlements. You could do it randomly, or based on terrain, or based on history. For example, if the 25 marches were once all part of a unified kingdom, and you’ve decided that March #17 was the old capital, then it would make sense to make that the largest village.
Question 3: Generally, the maximum level of the leaders is based on the realm population, yes. Refer to the chart on page 235. With 160,000 people in the area, the highest level NPC is probably 10th level. You’ll then have about 3 9th level and 8 8th level NPCs, while most of the 25 marches is probably ruled by a 7th level NPC.
If you are assuming that this is a very dangerous, undeveloped area, you could decide you wanted the NPCs be higher level, to represent the greater challenges. (Think of it this way: Only in 1,000 people is an investment banker, but if you happen to be on Wall Street, 1 in 20 is a investment banker). This would represent higher level characters from the meta-realm (i.e. the rest of the world) gravitating to where the action is. I wouldn’t hestitate to put the highest-level NPC being 11th or 12th level with 1-3 10th level, 3-5 9th level, and 8-12 8th level if you need to.
That said, the classic Keep on the Borderlands has huge keeps with sizable contingents (200+ soldiers) ruled by 6th level fighters, so you needn’t feel that the March Lords need to be 9th level.

Thanks for taking the time to answer my questions Alex.
Turns out, there is a precedent for population density below 40 people per square mile. In the establishing a stronghold section of chapter 7, it says the maximum density per square mile for wilderness hexes is 4 families per square mile.
I was close, but I realize I should have stuck with my gut and went with 20 people per square mile. Now to adjust the stats so they are all consistent.
Also, when you put it into perspective (6th level fighter rules the keep on the borderlands), I don’t mind the majority of the march lords being 7th level. The demographics seem fine.
I want to say again, you guys have put together an awesome game, and I feel excited just figuring this stuff out. As for someone who doesn’t have a background playing games in the old school manner (I got into the hobby 18 years ago, so around 93-94), the only tenuous claim I have is that I started with my uncle’s Moldvay Basic Rulebook. I moved into 2nd edition when I grouped up with the neighbor kids, and from there, 3rd edition, 3.5, and so on. I guess what I’m trying to say is, I don’t have the foundation for this manner of play except through what I read on the grogosphere, and you book makes it so much easier for me to set up this stuff rather than do it on my own without a lot of guidance. Thank you again!!!

Thanks for the kind words, Talath! I’m glad you are enjoying the game and that the assumptions and mechanics are all clicking for you.