Homebrew PCBs

>Hi Alan, nice to meet you, I'm Richard, the OP. 
>The problem falls into section 3.  I'm making a PCI-104 board with a
>lot of high count surface mount parts, and I need at least 4 layers,
>and probably 6.
>  
>
  Nope, definitely not #3.  #2.  'Problem' refers to your 'unresolvable 
problem' within Eagle.  Your problem is easily and clearly marked as is 
the solution, send in $400 for the Layout and Schematic license.  To 
short circuit the loop,  "I don't want to send in $400, or I don't have 
the $400 to send.."  Fine, that is your choice and is ok.  But in no way 
shape or form does it actually constitute an unresolvable problem within 
Eagle, as you characterized it in your OP..  It has limits, but they're 
clearly marked, and wanting to do something that goes beyond them 
doesn't even really constitute a problem with Eagle, and isn't unresolvable.

  Point being, Eagle is relatively solid and has extremely few real 
problems vs its large array of functions.  Someone taking your advice at 
point blank value is about 100 times more likely to hit a far more real, 
unresolvable problem in what you're suggesting as an alternative.  While 
it may be better if you want to do a free 4 layer board and aren't using 
advanced options, or some other rather specific thing, it is rather not 
likely to be the superior overall choice for what many people do, 
especially if it's still in some teething phase.  Once you get beyond 
learning the interface, you hit very few problems within Eagle, and 
almost any are easily worked around due to the nature of how Eagle is 
built at the lowest level.  There are plenty of valid gripes with Eagle 
and the interface, but hinting like some mysterious problem is going to 
stop people cold if they start using it isn't valid..

Alan