I've only done a few stalls in the -400 but this is my impression, and I bow to those who have done many.
The 747-400 has a benign stall charateristic, and reminds one of a Piper Cherokee. A bit of sloshing around and soggy controls and a floppyness in the air that was well described by my instructor as "like an old woman sh*gging." No doubt one day I will be able to compare the metaphor.
There is no marked pitch down, indeed I noticed no pitch down at all, but the surprising thing is that all the controls remain effective deep into the stall. So it is the matter of a moment to level the wings and lower the nose. The one aspect that is very surprising is that as you are lulled into the gentleness of the stall and its controllability you look at the VSI which is indicating in excess of 2500 fpm - and sometimes appreciably more.
So a nudger or a pusher is not needed, but if you ever get close to a stall in the -400 I would recommend that you get out of it mighty quick, as the altitude loss will surprise you.
I find our simulator model for the stall to be an adequate simulation but the physiological inputs which are so vital to stall recognition, such as sound and G variations are not realistic.
Yes, the Trident and the VC10 had a pusher, (VC10 less violent that the Trident) and they worked well as long as you did not isolate them on the very occasion that they worked properly. As many of you know the Papa India Trident crash had a contributory cause in that the stick push was isolated, and therefore could not do its job. Isolating the stick push was practiced at yearly intervals in the simulator, on the assumption that every fire of it would be false. Until the one occasion that it was right...