There is a limit on the number of bytes you can put for a PARM= in the JCL.
Try putting the filenames in a sequential dataset in the order you want. One
DSN per line. Then in the PARM field use the DD: specification instead of
listing all the dsnames. I don't have access to the mainframe right now, but
this is how I have it set up for our disaster recovery job streams.

Bill Boyer
DSS, Inc.