I am not sure Kublai Khan only has one source of horses. Enough spears and we don't have to worry much about Keshiks anyway.
Anyway, I will trade Writing with Kublai Khan when he will trade either Masonry or Meditation. I like to trade techs whenever practical and possible, as long I am not trading away techs that are directly dangerous.
Well, here's the dotmap that I promised:
Brown lines mark the fat crosses of the existing two cities. There's not much to say about the fish city (Site A) except the question of whether it should be settled next.
Site B is the ideal copper city site and can be considered quite important for our military because of its considerable early productive capabilities. It can help to pump out axes quite well when it's time to fight. Should we settle it soon?
Site C claims the horses in the west. There are two possible locations (C1 and C2), but I am leaning towards C2. Well, decision depends on whether a two coastal tile overlap with Djenne is better than having 3 ocean tiles in that city's fat cross.
Site D is nothing great, but it fills up the gap between NE city and capital, one that the AI might send a settler in a galley to fill. Overlap with capital is very unpleasant to me, though, as it contributes to a total of 7-tile overlap in Timbuktu if the dotmap is to be followed exactly. D1 seems way better because D2 would include another peak and a desert and moves away from the river. Either way, this city would have no food source except the sheep.
Site E seems to me the most plausible place to claim the silk and spice to the south. It has river in its fat cross but doesn't have access to fresh water and has hills around it to frustrate early farming efforts (until CS probably). If we want this site (i.e. before the war) we may have to settle it pretty soon before Kublai does, and that is probably some time before we get Calendar.
Finally, Site F is probably the best spot for claiming both the stone and the corn. It won't have access to either of this while Turfan still stands, though (hence the dotted lines to indicate its current impracticality), and its location would leave the lake (marked 'X') unworked, which is a bit sad for a Financial civ.
Any opinions? And what do you think should be the order of settling these sites? I am not sure we can get them all, but it helps to formulate a plan. I am tempted to stop at Sites A,B,C,D and then build up for war, especially if Kublai puts more cities near us.