Before this gets out of hand, let me clarify this, there's not such thing as more than 100 percent efficiency. If switching were perfect, you'd still have some amount of loss by the way resistive loss and hysteric loss (field collapse) in any PSU. There's always loss, it's a fact of life.
When i said 100+ percent, i think i meant either regulation, or feedback. 100 percent independent regulation means that each branch is regulated with it's own inductor and switcher combo and it's own feedback mechanism. There's another method of regulation, which i cited as 2.5 times more accurate and stable under load, or 150 percent more, as cascaded regulation, meaning the regulation loop is like waterfall whereby each regulation section dampens the one downstream from change. This only applies to the Turbocool series and cascaded regulation results in lower efficiency for that matter, but higher buffering and regulation accuracy.
Another analogy i might have used the extra percentage is feedback. It generally exists in all PSUs, but i believe it might have been misinterpreted. When a PSU's voltage is out of VREF range, or the reference voltage generated by a reference source or a zener (the PSU uses VRef as a 0 point), the error amp senses it due to it's comparator action, and thereby swings in the positive or negative direction to compensate.
The next part of feedback is more difficult as it uses Pulse Width Modulation which can only be explained in a duty cycle percentage, i used the analogy of a PSU being held under decent load as using 100 percent duty cycle as a baseline, when the load increases, the voltage drops and the the error amps senses it, thus it sends it's signal to the PWM generator to increase the duty cycle to say, 140 percent to compensate. Thus yes, duty cycle when compared to different base lines can have more than 100 percent (although the correct way to calculate duty cycles is Ton/Ttotalx100).
Sligh error in understanding i guess, no harm done. PSU design isn't easy, thus i feel a few errors in understanding isn't that hard to fathom. Heck i used to think switching PSUs used mini linear regulators insitu