With dual cam profile setups its not a case of the engine making low power off cam and high power on cam, its that each have an effective torque curve and you try and optimise both - if the ECU is changing over where the torque intersects then that is as good as you can get it, and if what Revhead said is true then that is the case. If with that setup you fall 500rpm below that point in certain gearchanges then the engines powerband is simply too narrow, or the gearing is not optimal for keeping it in that revrange, or perhaps a combination of the two.
On the flipside to this, is there actually a big step in power when high lift engages? If not, then again - I'd not stress about it.
As a practical example of why - with Jades Prelude I set VTEC to engage at around 6000rpm and in some shifts the car will fall under that point, so when caning it you can hear it fall off VTEC at times. This doesn't mean that its a problem where it engages, again all VTEC is for is maintaining a broad torque curve. I'm pretty sure you've seen his dyno plot, its basically a diagonal line going up and VTEC changeover is virtually indistinguishable... changing it from that point will make the car slower, so in his case (ie, with perfectly matched cams for the rest of his setup) falling off VTEC is by no means a problem beyond sounding a bit odd.
Being off high lift isn't the same as lag, by any means

Does that make sense?
Anyway, clearly its all too tricky and an old Corolla shouldn't get a motor with this fandangled technology. If it can run on pump gas and idle at under 1500rpm, you're doing it wrong - stay with the 4AGE and fix all the aformentioned issues.... valves are mean to be open.