On January 29, 1944 J. Presper Eckert wrote a memo explaining how information could be stored on spinning disks, using one large memory to hold intermediate results, lookup tables, and program information.  This was several months before von Neumann joined the project.  See A History of Computing in the Twentieth Century (1980).

von Neumann did a good job of writing up the idea, but it would have happened without him.

Andru