It's an interesting idea, but it'd take many years for solar-powered ion propulsion to push even a stripped down ISS out to a LaGrange point. Ion-propulsion thrust is pretty proportional to the amount of available electrical power. The solar concentrator arrays on the DS-1 probe produced 2500 watts at 1 AU. The thrust produced by DS-1's engine was about the same as that of a sheet of paper resting on your fingertips. Over a continuous thrusting period of about 20 months, that was enough to impart a Delta-V of about 4.5 km/sec on DS-1.

Even if the stripped ISS would have a mass that's a few orders of magnitude greater than the DSS. Suppose we could get 100 times the electrical power that DS-1 had. We could then generate roughly 100 times the thrust. We couldn't run the thrusters continuously because of the day/night cycles of the orbit, but we could go for about 45-50 minutes per orbit. At that rate, it'd take a long, long time to go where you wanted.

Some have proposed using ion engines to move the ISS to a lower inclination orbit, one more favorable to lunar and Mars exploration. However, as soon as you lowered the inclination by as little as a degree, then you've pretty well written off any further launches from Balkinour (I still call it Tyuratam).

There is something else to consider - just how undesireable is the ISS's current orbit when you're talking about sending spacecraft to the lunar poles? The really interesting places to go on the moon - the one's with the potential ice resources as well as the ones with continuous daylight, are all at the lunar poles. The Clemintine lunar orbiter was launched from Vandenberg AFB, CA on a Titan II booster, initially into a 63.8 degree inclination orbit.

By way of reference, the only other polar lunar orbiter that I can think of at the moment, Lunar Prospector, was launched from the Cape on an Athena booster. Obviously, you can launch a spacecraft into a polar lunar orbit from the Cape, but I wonder about the relative efficiency of going to the lunar poles from a 28.5 degree inclination orbit (from the Cape) as opposed to from the ISS's 51.6 degree inclination orbit. I suspect the answer is that it'd still be more efficient to go from a lower inclination, but I'd like to see some real numbers. If the penalty isn't too bad, having the ISS in its current orbit as an assemply point and staging area might not be a bad thing. The analysis would have to factor in the lowered capacity for launching something into the ISS's orbit in the first place as well as the energy required to then send something from LEO to the lunar poles.

Unfortunately, even if this were to work out for the Moon, it certainly wouldn't be the case for Mars. The ISS's inclination would cause a substancial penalty for Mars-bound flights.