I don't think you could have possibly achieved what I was trying to suggest, because simply adding a DIA isolation scheme should only impact which precursors (peptide + charge state or labeling) Skyline chooses. It should have no impact on transitions in the precursors chosen at all. That was the point. You would leave the range in your Transition Settings - Instrument tab untouched, as wide as you want if transitions are picked from library spectra that come from the same instrument, as the library spectra will not contain any peaks outside the instrument range. Though, if you are starting from a library created with your instrument, neither should it contain any precursors from outside its detectable range. So, this whole discussion may be moot, unless you are actually trying to limit your target selection in a way that is different from how the library was collected.
However, if your settings change impacts the transitions inside the precursors at all, then you are not achieving what I have suggested. I am suggesting that you temporarily choose DIA as you MS/MS acquisition method and then set a custom isolation scheme which ranges from 400 to 1000, as the only way currently in Skyline to limit the range of precursors isolated to something narrower than the range of MS/MS spectra on your instrument.
If you still can't make this work, please include screenshots of what you have done to the settings as it will be easier to explain relative to those.
Great to hear that you otherwise mostly got the method you wanted already.
--Brendan