How spam ACLs work

hi,

With respect to the exiscan smapassassin ACL, I have a few options:

warn (add the score)
warn (add scanned header)
warn (score over 5 add ***SPAM*** to subject)
deny (score > 12)

It works great.  But I need to limit the size of spam scanned mail to
80K.  Would I add an accept rule before all my 'warn' rules?  Or has
spam scanning already been done at thsi point? Does the scan take
place in my first 'warn' rule or does a scan take place for each one?

Can I use multiple 'condition = ' clauses?

Thanks,
Jason.