I flew and instructed in the T67 for several years. In general, a deliberate spin entry with full central aft stick and full rudder just before the stall induces a clean entry. Normally, centralising the controls within one and a half turns should result in recovery. (This makes it a good machine fo flick manoeuvres!) Care should be taken to ensure that the stick is moved far enough forward to put the elevators in the neutral position. This stick position is surprisingly far forward. Going back to the spin with correct pro-spin controls, a standard (CFS?) spin recovery technique is, once the direction of spin has been confirmed, throttle closed, full anti-spin rudder, pause, move stick progressively and centrally (aileron-wise) forward until the rotation stops, centre the rudder, roll to horizon, etc etc. The spin should recover in a couple of turns.
If aileron is applied at spin entry, the rate of roation may go high-speed a la Bulldog. Confirm that full normal pro-spin control is applied and ailerons neutral before recovery. The incipient stage recovery of merely centralising the controls may not work. (Differing inertia?)
IMHO, most 'scary spin' stories are the results of not taking/persisting with the correct recovery actions. The Chipmunk had similar scare stories in the 60s and 70s. Of course, without the tail-end mods fitted on most of the UK registered a/c and all the MoD fleet, the Chipmunk was more reluctant to recover than some.
One final point: the T67A has more critical longitudinal CofG limits than for its plastic siblings. These parameters should be carefully checked before spinning and aerobatics.
I recall there were problems with the spin charactersitics of the Tincanno when the A/B ratios were changed with the re-engined UK MoD fleet. Anybody care to comment re the 260hp version of the Firefly?