Every explosion has an overall energy release, and that release can be graphed in both the energy level of individual particle reactions vs. the volume of said particles. I suppose the energy release would be amplified by density, that is to say it would be higher towards the more 'compact' centre of the explosion (particle reactions against similar particle reactions).
I think the big bang was the same way. Most of the energy release into our universe was around the same level, but lessened towards the outer reaches of the 'explosion', meaning that the energy/quantity release (on the above chart model) would (also) look something like... one half of a bell curve.
I'm persuaded that this energy release resulted in an immediate particle formation, something almost exactly like a fermion-pair generation what we are able to observe today, but its epicentre was at much higher 'quark' energy levels. So, if a majority of individual energy releases were around the same level, their volume would drop off to either side of this 'energy centre'. I imagine that energies higher than this 'explosion level' would be much 'rarer', yet lower-energy releases much more 'common' and decreasing in strength with distance from the centre of the explosion... the graph would look more like a ladder leaned against an olympic ski jump ramp, wouldn't it? I don't see how it could have been any other way, as, if the energy release were all at the same uniform level, the fermion pairs would simply annihilate each other perfectly.
So, continuing this thought in this vein, what we would want to concentrate on is the highest volume of energy release: on one side, a sharp drop-off in higher-energy releases, and to the other, a more gradual decline (resembling one half of a bell curve). Now, imagine the very tip of this volume/energy curve. The energy release, even though it was simultaneous, would be in many different energy levels that are more and more mixed away from the explosion centre.
When we observe an isolated fermion pair formation, the particles annihilate each other almost immediately after their formation (as they are perfectly matched and the closest thing to each other)... but what would happen if there were (insert unimaginable amount) of simultaneous different-energy level fermion-pair formation around a single point? It is totally imaginable that the product of different fermion-pair formations could be closer to each other than their own 'genesis twin'. These 'closer' fermions, if opposing in charge, would try to annihilate each other instead of their respective twins.
So take this model, and apply it to a 'varied-energy' explosion... what if the two closest fermions weren't matched? They would still try desperately to annihilate each other, but since they aren't matched... I'm not so sure of the exact dynamics of this, but what seems most to make sense is that two opposing fermions within a certain energy bracket will partially annihilate each other, yet if it tries to annihilate another fermion outside a given bracket, it won't succeed, and be 'bound' to it in an eternal struggle (until later events disturb it).
Now, let's just concentrate on the dominant energy level release, and the one just down form its 'annihilable bracket'. What would happen just after the energy release (massive fermion pair formation) is that a given number of those fermions would annihilate each other, but another part would 'cross-bind' in the way described in the previous paragraph.
Now, we get into the dynamics of 'why matter?', because these leftover 'mismatched' fermion pairs would equal each other and be divided in respective charge - a negative higher-energy bound to a positive lower-energy, and this would have an opposite twin - thus these could annihilate each other, too, but that's not what happened.