Adding the higher capacity pump will help, it did in my case but there are limits. a 618 is a large body and the 1048 is about as wimpy as a pump can be, I would run a 1260 on it, the piping will limit the GPH but the 1260 is plenty strong. Virtual height is real and does have an impact. I would not change reactors, the GEO is a good piece of equipment, but it can be improved upon. For reference, I use a 1250 on a 3x12 (actual media holding capacity is 3x9) body and it is about perfect.
Borrowed from JDieck here...
http://reef.diesyst.com/crarticle/crarticle.htm
"Virtual height of the media column: What? Well let me explain. Every time the water passes trough the media it is equivalent to having a media column equal to its height. If the water re-circulates five times, it will be equivalent to have a media column five times taller in comparison to a reactor with no re-circulation pump thus achieving a virtual height of five times. A more powerful pump may re-circulate say ten times thus creating a virtual height twice as tall as the reactor with the pump re-circulating only five times.
The higher the virtual height of the media the longer the time the media will be in contact with the acidic water or in other words the longer the retention time will be and the longer the time, the more effective the dissolution and CO2 consumption will be.
Also the higher the virtual height, the higher the water velocity will be thus making it more effective to drive the calcium and carbonate ions away from the media surface resulting in an increased performance.
In summary, a larger re-circulation pump will increase the performance by increasing the virtual height of the media thus the increase in retention time and the flow velocity.
Now hold your horse! This does not mean that installing a swimming pool size pump in a reactor will make it better, of course there are practical and physical limitations to the size of the pump, limitations like wasted power consumption, propensity for cavitation at the pump, vibration, media tumbling and breaking apart, media carry over and so on."
Some may not agree, and that is ok, my own experience shows me that it does make a difference, as well as adding a gas recirc loop. When trying to crank the reactor up, not all of the CO2 gets dissolved and collects in the highest part of the reactor, in most reactors this is where the effluent leaves the system, and also where any undissolved gas exits, and you don't want that, you want as close to 100% of the CO2 to be used, not expelled. GEO's are not as bad as some reactors with expelling gas, but they can. That probably isn't you issue you are having so I would concentrate on getting rid of the wimpy pump.