I think when Neil says "useable" he is referring to glazes which have silica, alumina and flux amounts which make them suitable for functional surfaces according to general accepted guidelines. Is that what you mean Neil?
I am sympathetic to Neil's desire (I would love all my glazes to be functional!), but would argue against this kind of modification of the currie technique, at least to start out with. Limiting the grid corners means you miss out on all the wonderful decorative (but non-functional) glazes which can magically emerge from the currie technique. Some of the most interesting glazes I have found have overloaded one ingredient or another. You would not feel comfortable putting them on an eating or drinking surface, but they still work in the sense that they will melt on the pot - and look great! I am not advocating putting just anything on your pots - there must still be the usual potter-beware judiciousness in place, but that still leaves infinite possibilities.
The other wrinkle is that to vary Currie's standard grid (ie, change how his existing calculations page works), you need Matrix glaze software (which has a currie module) or your own custom calcs or spreadsheet. Insight does not do it easily, although I think what Neil is implying is that you can go plug in recipes by trial and error and eventually figure out what blends are inside the limits, but this might take a while. However, I like Neil's thinking about using the grid axes to vary fluxes (not just clay/alumina and silica).
Another reason to use Currie's technique as it is out of the box is that you can learn a great deal looking at "extreme" blends of your glaze components. I have a glaze I am working on at the moment which was not as glossy as I wanted - something in the glaze was leaving a very subtle kind of swirling roughness on the surface, even in cells which fell within functional limits. It was only by looking at extreme blends in the silica heavy part of the, and then visually following the silica trail down the grid, tile that I realized that this roughness I was seeing was due to a still ever-so-slight silica overload!
Having many slightly differentiated tests of a glaze mixture on the same ceramic object, fired in the same firing, in the same place in the kiln makes those results highly comparable and the learning from them is increased. Since I can get a very good indication of melting, runny-ness and vertical movement of a particular glaze from the currie tiles themselves, I wait to do individual tile tests until I cherry-pick the grid for what I want to test further. And I can say, hand-on-heart, that more than once I have gone straight from the grid results to mixing up a much larger batch for use on pots, without going through a second-stage test tile run - although this is not without some risk.
I also find that once I have used the same currie batch to make up three or four grid tiles from for testing on different clays, reduction vs oxidation, etc.. that I don't really have enough of each glaze left in the cups to properly dip test tiles.
I don't think there is really any wrong or right in the experimenting stage (says the guy who is flying by the seat of his pants with native materials), as long as you are cognizant of the risks, and realize the strengths and weaknesses of your results. As you can see from this discussion, this technique is very powerful, and can be used in SO MANY different ways.