Honestly, I'd personally accept the 4+4, IF the subject of DOO is removed or at least clarified. Compulsory redundancies could be changed to voluntary and would (I suspect) be over subscribed.
Those advocating for DOO do realise that ticket prices wouldn't decrease if the second member of staff was removed don't they?
The Treasury treats railway costs and revenues as completely separate to each other; prices are increased each by a certain amount each year (RPI plus or minus whatever percentage is politically convenient) regardless of what happens to costs.
There are three main benefits to DOO, depending on the way it is implemented.
Firstly, if the driver releases and closes the doors, there is a time saving at each stop, as there is no need for wait for the guard to undertake these tasks and give two on the buzzer before the train can depart. Some operators already have driver release and/or allow guards to release doors as soon as the train comes to a stop, but there are still many operators/traction types where this is not the case.
Secondly, if the guard is no longer safety critical, then the train can run without them. This is, for example, the case with ScotRail Ticket Examiners and Southern OBSs. This means that during periods of disruption, or if a guard calls in sick, trains don't have to be cancelled for want of a guard, so reliability is increased. This also means that spare coverage for guards can be reduced or even eliminated entirely, and that staff don't have to receive as much initial or ongoing training; these both present moderate savings.
Finally, if guards are abolished entirely, there is clearly a much larger saving to be had - even if you don't have any compulsory redundancies. However, it seems very unlikely that this will happen.
Whilst the biggest financial savings come from the second and third changes as described above, driver release/close would mean significant benefits for dwell times, increasing reliability across the country without needing to change the timetable. Even just driver release would have an appreciable benefit.
I think the most that can realistically be expected is for driver release, and possibly driver close, to be introduced. The former is something which almost all traction types are already capable of doing, and generally doesn't require any additional infrastructure (lighting/CCTV monitors etc.). At some stations (such as where local door operation is used), guard release would need to be kept - but this represents a minority of stations.