You could ask the same question of calculus (or any theoretical domain of knowledge). The rules of derivatives and integrals don't, in themselves, give a model

You could ask the same question of calculus (or any theoretical domain of knowledge). The rules of derivatives and integrals don't, in themselves, give a model of any particular empirical system. But they are a tool/language that can be used to describe/build/evaluate models for systems. To give examples requires an agreement on exactly which domains of knowledge you consider to be part of "complex systems", so based on your question I'll assume that you consider chaos to be part of it. For example, the Poincaré-Bendixson theorem saves researchers time because they don't have to consider the possibility of chaos in 2D continuous state space models. For example, since we know that weather is a chaotic dynamical system, we can give up on the hope of ever being able to precisely predict the weather a year beforehand. This saves researchers from spending their time on impossible objectives.