The AMS-02 collaboration released several secondary-to-primary ratios of
unprecedented accuracy. These ratios can be used to test the
universality of propagation for different species, and also to test the
presence of breaks in the diffusion coefficient. It was shown in Weinrich
et al. (A&A 639, 131, 2020) that the combined analysis of Li/C, Be/C,
and B/C strengthens the case for a low-rigidity diffusion break. It was also
shown that a standard propagation model successfully reproduces these
ratios (and also AMS-02 N/O and 3He/4He data), without the need for
additional sources of Li, Be, or B. However, significant modifications
(~5-15$\%$) of the production cross sections are required, though these
modifications remain within estimated nuclear uncertainties. We also
extend our analyses to the recently published F/Si ratio and discuss
how much F at the source can be accommodated by the data.