Some thoughts on what might be done on VBF studies:

There are three programs on the market:

HAWK (formerly CDD) and VBFNLO (Zeppenfeld et al) and VV2H (Spira). They have 
done the comparisons for the total cross sections at the 2007 Les Houches. 
HAWK is now also available interfaced to LHApdf, so it would be nice to look 
at the differences between MSTW2008 and CTEQ6.6.

Another thing is a about the electroweak corrections. These are in HAWK, and 
probably soon also in VBFNLO (though they just used the results of the CDD 
paper). So it might be useful to get some new comparisons done. This might 
also be extended to differential distributions, but here VBFNLO gives more 
oppurtunities, since it also has Higgs decays.

Another thing that might be very useful is a true NLO calculation, matched 
to a parton-shower. This is implemented into POWHEG, but I briefly talked to 
Stefan Dittmaier and he does not think that all the corrections are in there. 
He and his collaborators are themselves working on NLO-PS-Matching for VBF, 
but this might be a more long-term project.

There is the experimental aspect of getting extra jet activity between the 
tagging jet. Especially the Karlsruhe group has done a lot there (Kristov 
Hackstein is the CMS contact..).

Another thing is the proper treatment of finite Higgs widths - this becomes 
more important for high masses, but is needed only above 300 or 400 GeV or 
so. I do not think it makes a lot of sense to look into it as a first thing, 
but instead concentrate on the smaller masses.