Original poster: Steve Conner <steve@xxxxxxxxxxxx>
I would love to see a photo of something that is better than the 
trifilar wound toroids that I referred to.
James Pawson did a lot of work on this a couple of years ago and 
claims leakage inductances as low as 0.08 microhenries with about 
450uH magnetizing inductance. The datasheet you posted claims 0.47uH 
with about the same magnetizing inductance. So amateur built GDTs 
can do about 6 times better than the LSE part.
http://thedatastream.4hv.org/gdt_prac.html
I don't know if the multifilar screened wire pack is a new idea. I 
have seen baluns made with coax but never multiple magnet wires 
stuffed inside a braided tube. If it is new, hopefully it was James 
or me who invented it ;-)
Steve Conner
http://www.scopeboy.com/