You could add a few paragraphs at the beginning of the documentation to explain the general structure of the code and the terminology. From the example, I am guessing from the examples that pfsspy.pfss is the main function solving the PDE problem and that "tracers" are post-processing tools to reconstruct the values of the magnetic field from the abstract output of pfss. Maybe this could be explained more explicitly. By the way, wouldn't solve or solve_pfss be a more explicit name for pfss?
From https://github.com/dstansby/pfsspy/issues/252: