First thing to do is search:
http://srforums.prosoundweb.com/index.php/m/37751/4978You'll find the links in most of my posts on this thread busted - I still have the pictures if you're interested but for good reasons the mods have disabled the ability to edit older posts.
The question you've asked is huge and complicated. You can't just try these filters, you have to have this filter result in the acoustic realm for it to work, i.e. electrical filter + raw driver output in cabinet = acoustic output. I'm assuming the amp has a flat transfer function with varying load impedance - not all of them do.
The whole point of the Thiele filter is to achieve a steeper cutoff slope without the audible phase and ringing problems associated with classically formed filters exceeding 24dB per octave. This filter is another great idea from a guy that the loudspeaker industry already owes a great debt to. It basically allows you to increase the cutoff slope in such a way as to have a benign affect on phase at the expense of an inaudible bump lower down on the "skirt" of the filter. Remember this has to happen in the acoustic domain and is tough to pull off correctly. Using that BSS processor on a real loudspeaker without a bunch of other parametrics isn't going to work and likely to give you a worse result than a standard 24dB Linkwitz Riley filter.
I know one guy that pulled it off very nicely with EAW's KF730 in the low bandpass. Interestingly he did it in both the MX8750 processing originally recommended for this loudspeaker as well as in the more recent UX8800. He used a notch filter added to a standard low pass filter (see graph below). As you can imagine, this type of filter can be implemented with many other processors by adding a the right notch at the right location, but if your name isn't Neville, Dave or Tom you've got your work cut out for you.
If you really want to get into this area of loudspeaker design I'd look into high slope linear phase FIR filters.
