Point spread functions describe diffraction and scattering in imaging instruments and can be used to correct the images for the scattered light. However, mostly, the scattered light correction is not used in the solar physics community, on the one hand since many scientists are not aware of its importance, on the other hand since it can be computationaly extensive. In this talk, I will present
new revised point spread functions for AIA that correct for long-distance scattered light over the detector,
what solar features and data products are affected the most by instrumental scattered light,
the error involved when the scattered-light correction is neglected,
and a new point spread function deconvolution method, which is faster by a factor of 4 to 50 compared to the Richardson-Lucy algorithm.
I would like to conclude with a discussion on what further data products and scientific results can and should be improved by correcting the underlying imaging data for instrumental scattering.