Well yes... its very, very common to se eye rigs where you have two locators under frouped under one then you can scale the damn thing to look at something thats close andscale up to keep the eyes looking at a distance so that the locator never has to travel wery far from the eye! Thisway you can avoid the problem you have and still have the benefits for close up precision if needed!
I animated a character set up like that once. The two individual eye targets were in a group aim-constrained towards the head, to keep them aligned.
But it had a quirk: when the character turned his head sideways (which of course happened repeatedly in my shots -- after all 3d animation is basically a living dissertation on Murphy's Law) the two eye controls would remain in a horizontal configuration, causing the eyeballs to sort of 'torque' in the head; one pointing up and the other down. I had to keep counteranimating the individual eye targets to stabilize the eyes. No fun at all.
In hindsight, I guess that problem would never have occured if the aim constraint had been given a proper up vector defined by the head. So perhaps my disdain for aim controls is partly irrational and just based on a bad experience =)
But for the time being I think I will stick with and continue to advocate the orient constraint solution; it's simpler than dirt and works right... traits of a good rig IMO. In the end, either system is more than sufficient to solve the problem. I guess it's a matter of animator preference.
