FOM: completeness theorem for stratification?

[Stephen G Simpson]

Thomas Forster (FOM, 13 Apr 2000):

 > a formula is (equivlent to a) stratified formula iff the class of
 > its models is closed under [ the Rieger-Bernays permutation ]
 > construction.  A bit of good old-fashioned model theory!

OK, very interesting.  And please do send me the offprints.

Where does this leave us with respect to my earlier question (FOM, 7
Mar 2000) about the set-theoretic intuition?  Is there a clear and
convincing picture which underlies NF, similar to how the picture of
the cumulative hierarchy underlies ZFC?

-- Steve