There should be understandable correlation between structural biology and physiological function, e.g., http://www.news-medical.net/whitepaper/20160315/Novel-use-of-EPR-spectroscopy-to-study-in-vivo-protein-structure.aspx
One thing you can do with NMR, but not with X-ray diffraction is show very transient protein-peptide interactions. I suppose that can be seen as a source of complementarity.