For example $\{\land, \lnot\}$ apparently forms a functionally complete set as it can form any logical expression.
But I don't really know what this means or how you show it. Do we just show that some set is complete and then in the future if we have a new set, show that we can replicate the known operations with the new ones?
Do you have to find these strategically or is there a systematic way?
Yeah, that's how we typically show that $\{ \land , \neg \}$ is complete: we already know that $\{ \land, \lor, \neg \}$ is functionally complete, and we just point out that $\lor$ can be rewritten in terms of $\land $ and $\neg$:
$$P \lor Q \Leftrightarrow \neg (\neg P \land \neg Q)$$
Also, here is a post explaining why $\{ \land , \lor, \neg \}$ is complete