My interpretation is that if the galaxy uses 425 watts, its most likely driving the bulb at about 400 watts. With the PFO HQI/HPS430, the wattage used can be over 500 watts (536 watts with the AC 14kK)... so the bulb is getting something like 430 to 450 watts...
http://www.advancedaquarist.com/2005/1/feature/view
Your Radium SE bulbs are actually considered 270 watt bulbs, not 250's so much... because on a M80 ballast, thats what they get.
I was wondering if your current 250's are on icecaps or M80's though, because IME, the bulbs have a much 'fuller' spectrum on HQI ballasts, where on icecaps they are somewhat monochromatic blue.
When you look at the graphs of the spectrum, you will see the spectral plots are almost like someone is just moving the whole graph up and down, but the curve/plot stays the same. With blue-spike bulbs, this means that as you go up in wattage, you are raising the levels of 'daylight' by a hefty percentage. On an icecap, the bulb may have little to no red-orange at all, but on a HQI, you might see an increase of 100% or more, simply because there is so little to start with... where the blue spike seems to shift up just as much, this is really a small percentage of blue... maybe 5%-10% at best. It seems like the spectrum is 'fixed', but when you compare relative percentages of wavelengths, there is a considerable difference in the way the bulbs look. Even though the 400 watt version of the AC looks like it just dominates in blue and actinic (even over the Radium, which it does), if you compare the other 'full' spectrums, there is a considerable hike in the levels of red through green as well. To your corals, this being 'less monochromatic' can be the difference between 'chalky washed out' colors, and vibrant ones. I hate the pheonix 14,000K on icecap/e-ballast for instance, but love it on HQI/M80. Its a totally different bulb, esp after a year of being run like that. On the M80, it looks more 'white', after a year and staring at it for a bit, you swear you can actually see red and yellow out of it. After a year on e-ballasts, the things seem to get very dull, and monochromatic blue... they look more like a 20,000K... and my corals started looking 'chalky'.
Depending on if you overdrive or underdrive, you will likely see the same relation on 400 watt bulbs as well. Its why some people still prefer to run the Radium on HQI ballasts, even though running just a probe start bulb on a HQI ballast tends to cause parts of the bulb to unweld themselves and cause some nasty failures. But in general, most 400 watt 20,000Kish bulbs tend to have more daylight... not just the monochromatic blue spike with a little actinic like alot of 250's or even worse, the 150's. Thats also how many bluer 20,000Ks actually start to regain a PPFD/watt ratio and some like the Aquaconnect can actually compete with 10,000Kish bulbs. Heck, at 1000 watts, 20,000Ks can compete with 10,000Ks.
But just look at these PAR levels recorded off of a 8x5' T5 (run on Icecap ballasts) array and compare to the composite I did of your tank. This is a 60x24x24 tank, and the white balance was screwey on the guy's camera... but those PPFD levels are just shredding!
Here were the levels from your tank:
So thats 3x250watt bulbs with 4x54wattT5's over a 180g (60x30x24) for you vs the T5 array above...
Although if you went with the 400's, I would say that dual 400's would most likely be enough... esp if you used dual lumenarc 3 full size units, although I know you are partial to SLS if nothing else for size reasons. With 400's you would see some slightly higher levels up top on your tank, but the spectral quality is what would change the most IMO.