TAT is not the determining temperature
JTR,
Sorry if I was not clear enough. In answer to your question, "Is it possible for the fuel to get colder than TAT?" - the answer is sure, why not? The point of my earlier posting is that the wing skin temperature has got very little to do with TAT, and everything to do with SAT. The temperature of the stagnation point on the leading edge might be TAT, but that isn't where you are storing your fuel. The temperature of the air over the wing skin is a function of SAT and aerodynamic lift, and in particular is COLDER than SAT (which itself is of course a lot colder than TAT). So, in your example with a cruise SAT of -70, the airflow directly adjacent to the wing skin would be colder than -70. I can't tell you how much colder because that's a function of local fluid dynamics, although you might be able to do a back-of-the-envelope guess by figuring out how much "suck" it takes, divided by the wing area, to hold your megajet up in the air, and then making a few broad assumptions about relative pressure between the top and bottom of the wing.
Even if you assume that the wing surface airflow is at SAT, since you are saying that after a long cruise the relative temperatures reach a steady state of SAT=-70, Fuel Temp =-40 then it sounds like your heat exchanger is able to maintain a 30 degree differential. All other things being equal, and taking a large pinch of salt with this, if you lost half of the heat source, you might reasonably expect to maintain 15 or more degrees of differential. (More, because heat losses would reduce as the fuel temp approaches SAT). So - and again I emphasise this is back-of-an-envelope stuff - you might expect to see perhaps -50 to -53-ish if you lost half your heat.
In practice, you may not lose half your heat when you shut down an engine. I have no idea how the system is set up, but you might find that with one shut down you actually get more heat from the remaining engine than when both are running, so that the temperature is not as bad as -50. This is right around the freezing point of Jet A1 by the way, although well below the freezing point of Jet A (which can be up to -40). Since you are already operating with the fuel at -40, I assume you are on Jet A1 already.
Finally, I know that Boeing have a software product called the Fuel Temperature Prediction Program for the 777 that models wing airflow temperature, insolation, friction warming, and contributions from the heat exchangers, fuel pumps, etc., to allow operators to plan things like polar routes. I don't know whether you have a version for your type, but just to confirm all the above waffle - in the FTPP examples I have seen, in routine ops (not OEI), the fuel temperature falls below TAT (but stays above SAT).