When you do "Apply peak to subsequent", new peaks get chosen for all of the samples that were acquired later than the replicate that you clicked on.
Skyline reads the acquisition time from the raw file when you extract chromatograms.
If Skyline is unable to read the acquisition file from the raw files, then "Apply peak to all" would behave the same as "Apply peak to subsequent", since Skyline would not know which replicate came before which.
You can see what Skyline thinks the acquired time ordering of your replicates is by right-clicking on the Peak Areas Replicate Comparison graph and choosing "Order > Acquired Time" from the right-click menu.
You can use the document grid to see what the Acquired Time is for each of your replicates:
View > Document Grid
Choose "Replicate" from the Views dropdown, then choose "Customize View".
You can find the Acquired Time column in the Customize View dialog at:
Replicates > Files > Acquired Time
It is possible that Skyline is not reading the acquired time from your files. If you think that is happening (or if anything else weird is happening), you should send us your files and we will see what is going on.
In Skyline, you can use the menu item:
File > Share > (complete)
to create a .zip file containing your Skyline document and supporting files, including extracted chromatograms.
You can upload that .zip file here:
https://skyline.ms/files.url
It would probably also be helpful if you uploaded at least one of your raw files too.
Hope this helps,
-- Nick