Hi Max,
Once you have your 'suffix -c' images (these being a collection of your 'Calibrated Lights') you then need to 'feed' this collection into the StarAlignment process - perhaps then creating a new set of 'suffix -c -r' images (these now being a collection of 'Calibrated and Registered Lights').
With your -c-r images, you then 'feed' these into ImageIntegration, eventually creating your 'MasterLight', ready for subsequent processing.
Obviously, if you are working on a mono camera with separate filters, then you will be creating a 'MasterLight' for each filter used during the acquisition process. If this IS the case, then it can be beneficial to pick ONE image from all of your data to be the 'reference' image for the StarAlignment process (this just means that you will apply fewer, un-necessary, image transformations - hopefully resulting in less 'noise injection' during the pre-processing phase, althouigh the 32-bit processing power of PI, coupled with Juan's impressive data manipulation algorithms means that you probably wouldn't see the effect anyway

)
HTH
Cheers,