> As much as possible.

That is not a useful response. Would it be worth doing if "possible" turned out to be 0.1 mW? 1 mW? 10 mW? 100 mW? 1 W? 10 W? 100 W? 1 kW? 10 kW? Presumably somewhere in that sequence your answer goes from "no" to yes" and that point determines what tradeoffs you're going to be willing to accept in order to increase your power capacity.

People have suggested steam engines. Those would definitely produce more power than TECs, at least twice as much and potentially six times as much. But they are far more likely to kill you. Is that tradeoff worth it to you?

TECs are pretty expensive per watt. Are you really willing to spend tens of thousands of dollars if it will increase your energy output a little? How about hundreds of thousands?

There are tradeoffs in any engineering design. It's obvious that more power is better, but without some idea of the shape of your utility curve, it's impossible to evaluate those tradeoffs in a useful way.