SeqMonk can only store one probe list so if you regenerate your probes then the previous list is lost. If you therefore want to be able to make some comparisons between different ways of generating probes then you need to create parallel versions of your project.
An alternative to this is to convert a filtered probe list to a feature track before generating a new list. In this way you can preserve the positions of regions of interest and use them either visually or to annotate future probe lists.
The ProbeList annotation parser simply converts the currently active probe list into an annotation track. Both the annotation set and the individual features created are named after the probe list you choose by default. You can opt to change the type of features created when you make the new annotation set. The name of the annotation set can be changed later, but the feature type cannot.
There are no options to set so as soon as you select this option you will see a new annotation set appear in your data view and the new track will immediately be available in the annotaton track selector.