My plan for getting my economy & research back in order is...
I would take a different approach to an economic recovery plan. Your economy isn't weak because of lack of trade routes or high city maintenance. The problem is lack of population, and that's due to lack of

. The best thing you can do short term to help your economy is raise the happy cap, and that means Monarchy. If you go to 100% research you would be able to finish Meditation->Priesthood->Monarchy in about 22 turns, which just happens to be how long your gold will last. Of course any economic improvement will shorten the research time, although increases in expenses will also decrease the number of turns before the gold runs out, but you should be able to get it done before research has to be turned down.
Once HR is in place your cities have enough food to grow quickly. Gondar can work as many as 17 cottages (eventually) and even Lalibela can work 10. So raising

will lead to a much stronger economy faster than any other approach.

-wise you're in good shape, +8

from resources (w/ Granary), with one more

coming from Lalibela's deer. So cities can reach the mid-teens before

will become an issue. Even though it will hurt your economy you're going to have to continue to build military over the next round to avoid a major power drop, so you might as well put them to some use.
It won't be an issue if you decide to pursue Monarchy, but I disagree with one of your statements:
Meditation -> Priesthood -> Code of Laws is certainly the cheapest, fastest path to CoL, and not one I usually follow.
I'm not sure if that's what you meant to say, but if it is you might want to rethink this in the future. Because of the research mechanics, researching CoL without knowing Priesthood is very questionable strategy IMO. Because of the major

cost difference between CoL and Meditation + Priesthood you should get back most of the

s you put into the the cheap religious techs because of the extra 20% pre-req bonus you'll get. You might be thinking that you can get them for free at some point via trade, but that means having something to trade (which isn't guaranteed) plus it can have WFYABTA implications.
Another research issue is what to do with the GS due in Babylon in 17 turns. Settling or an Academy are viable options, although I don't see an obvious location. Currently the GS would bulb Math, but if you hold onto it until you research Math yourself Alphabet is next.
As for the GG, I would normally look to settle him in these circumstances, since having a Medic 3 chariot sitting around for 50 or 100 turns before your next war seems like a waste. It looks like Aksum and Babylon have similar

potential (Babylon a little better short term, Aksum a little better long term). Since Babylon has much more food I would think it would be best to make Aksum your HE city (unless something better comes up), so I would settle the GG there. Of course that does raise the issue of whether it makes sense to cottage spam Aksum.
Regarding military matters, it seems that you've skimped on your barb defenses while taking out Zara. Gondar looks especially vulnerable at the moment. You're going to want to leave a significant military presence in the north in case Saladin gets any funny ideas, but don't forget the threat from the south and east. For exploring in barb spawning regions you might want to consider using an axe instead of a chariot. If a chariot ends its turn next to a barb spear or axe it will probably die. Although exploration will take longer, an exploring axe should have good odds against any barb it comes across (especially if you give it Shock).
Finally, I think you'll want to explore some more before deciding on city sites to settle, but I'll point out that the hill 3S2E of Babylon looks like an attractive spot. At pop 11 it can stagnate working the lake, 7 grassland cottages and 3 mined (or cottaged) grassland hills, with none of the tiles shared with other cities. It would miss out on the sheep, but they'll be available to feed a city along the coast to the south. But more exploration might reveal better options.