I think it would be useful to be able to supply PI with an sample for measurement of actual noise for a given workflow/optical/digital/etc. context, which would then replace inbuilt/default noise "assumptions" used in various algorithms.
For example in Restoration, the Wiener deconvolution algorithm utilizes a noise term which may or may not match the actual noise in the context. It would be a useful option to be able to provide an image or select a rectangular portion of the image to designate as noise.
