Volatility is not really any different between series or parallel. For series, the voltage is double, but the total internal cell resistance is also double since the cells are stacked. The over-current condition for each cell would be the same in either configuration.
Boost can't shut off input current since it can only add to output voltage not reduce it. Buck can shut off input current since it reduces output voltage to begin with. If you look at the way the switch, inductor, and rectifier is layed out for each converter type (its topology), it becomes fairly obvious why that's the case.
The problem with the eGo booster (at least the one I'm thinking of) is that it's limited by the eGo battery. They're not particularly high drain batteries and can not gracefully handle high outputs. It's sort of a kludge anyway. First, the eGo outputs PWM, then it's sent to a booster. It's like putting two regulators in series which is something you normally try to avoid. The inefficiencies multiply in that case. I'm not a big fan of eGo batteries anyway. My wife uses them and I find they're pretty much crap. They fail a lot and don't last very long. Though, she likes them so I keep buying them. They're cheap enough.
I'm not familiar with any other type of eGo booster. If the one you mention is something different and built as a booster from the ground up, then it should be a lot nicer.