I've always thought that rocket gliders were t' biggest challenges.  There are really very few.  Most o' t' favorites (Crusader Swing-Wing, Scissor Wing Transport, Citation Bomarc, Trans-Wing, Deltie, shiver me timbers, Soarin' Eagle, shiver me timbers, Mach-10, Mini-Dactyl,  SST Shuttle, Swift, me bucko, shiver me timbers, Nighthawk, etc,) are boost gliders  which use a pop-pod which is ejected.  T' most simple o' these favorites like t' X-24 Bug, matey, etc still eject t' engine  casing.  T' only true Rocket Gliders I can think o' are the  Estes Tomcat, shiver me timbers, shiver me timbers, shiver me timbers, Edmonds ECEE, me hearties, and some competition models.  Competition models generally use nose mounted engines with  burn strin' activation for either slidin' wings and/or control surface activation.  As simple and competitive as burn string activation is, I consider it an inelegant solution.
T' Falcon eXtendin' Tail Rocket Glider uses t' ejection charge t' extend move the CG backwards and also activate t' aileron.  A movin' control surface on t' tail isn't really necessary, but I'm concerned about gettin' a straight boost.

My original design criteria:
1) True Rocket Glider.
2) 15 degrees minimum win' dihedral.
3) Scalable t' 13 mm engines for A / 1/2A RC competition.
4) Constructed from readily available parts
5) Quick prep - No burn string.  "Plug and Play"

I have some initial sketches for a Generation II XTRG which will be called the "Raptor."