Hi Paul,
Yes, The way I understand it is if you are using the AOD function in ISIS to calculate and correct for atmospheric extinction then the resulting instrument response just takes into account the instrument (ie not the “IR+extinction” commonly calculated using a nearby reference star.) You are therefore correcting your spectrum in two stages, instrument response and atmospheric extinction. (I understand this is the usual technique used by professionals who know their instrument response and therefore just correct for the atmosphere for that particular observation) You can see an example of the procedure here.
http://www.astrosurf.com/buil/atmosphere/annexe.htm
EDIT: linked from
http://www.astrosurf.com/buil/atmosphere/transmission.htm
Cheers
Robin