I am curious about the wavelengths which both John (implicitly in the ISIS listings) and Robin give for their calibration lamps. These are different from the standard Ar and Ne wavelengths given by NIST, for example here https://physics.nist.gov/PhysRefData/ASD/lines_form.html
If you look at the discrepancies which ISIS lists in John’s calibration fit, these appear to be almost exactly what is needed to bring the lines into agreement with the NIST values.
I have the Shelyak Ar-Ne lamp in my LISA and use a .lst file in ISIS which contains the NIST wavelength values. This routinely gives me a 4th order polynomial fit with rms residual less than 0.1.