I'm not convinced there are any major balance issues with the missiles vs beam argument, but I would like to give my thoughts on Paul M's points;
1) Why are fire controls so large? At the ridiculously long ranges we're talking about for beam and ballistic weapons you not only need to determine where your target will be at the time your weapon will impact it, but also need to account for the light speed lag associated with attempting to view something so far away, now if we assume that transnewtonian physics allows objects to be tracked near instantaneously without any relativistic lag, perhaps the equipment necessary for such physics bending are rather cumbersome? As pointed out somewhere else discussing this subject, the angle necessary to miss an object at such ranges would at some point approach infinitely small. Perhaps the fire controls also aim the beam as it travels using gravitational lensing or something. Where as tracking a missile from such range I assume would use more active technology, you guess where the object is based on the ghost on your sensors but as the missile gets closer the error associated with your fire control gets lower because of the lower range.
I don't really agree that fire controls are too large compared to beam weapons as for fighters you can reduce their size and range or tracking speed, fighter sized ones already get a tracking speed bonus which helps immensely, but for larger craft the fire control gets even less important as you pile more weapons onto less fire controls.
2)I think the size of missile fire controls is about right as active sensors are rather unwieldly already, but I think Paul's right about ECM/ECCM being somewhat unbalanced, It's simply too easy to overbuild fire control range, perhaps ECM/ECCM should have a random chance involved with the effect it has on your fire control and also effect active sensors in the same way.
Perhaps a good balance would be that each Level of ECM adds a 70% failure rate to enemy fire control untill it reaches maybe 10%-20 it's designed range, equal matched ECM and ECCM reduces that to 1%, Varying levels of ECM and ECCM would either add or reduce the error rate, I haven't worked out a good equation for it, but the rate should be high when theres more ECM and low when theres more ECCM. The idea of error rate is that when the sensor is checked to see what's in range a roll is made against each contact found based on ECM/ECCM, if the roll fails the contact disappears, but the roll must be modified by range so that the effect diminishes significantly. The effect of this would be rather more interesting than our current system, basically ECM shielded fleets would be hidden from your sensors untill they've crept somewhat past your sensor range, and while they were still on the perimeter would alternately drop in or out of view, missiles would need to keep travelling towards the last known contact without actually having any active sensors otherwise it would disrupt the current system, but when the fire control requires the target the missile changes heading, If the target vanishes off the fire control or sensors for some significant time before being reaquired then significant fuel would be wasted as the missile attempts to change heading for it. Furthermore if the contact vanishes as a missile is about to close on it it would obviously miss and at this point active sensors would be needed for it to reaquire the contact. Finally if the fire control of the ship that fired the missile is destroyed then as the current system the missile would self destruct if it has no actives. I would add that nebulas and black holes should increase the base error rate.
3) Agree absolutely, pre version 6 had rather logical fuel useage( for missiles and fighters that is,
I think large ships seems about right)
4) I'm in agreement that the hit percentage being mostly based on speed might not be the best, if we assume that Aurora's missiles aren't kinetic since more speed and mass don't increase warhead size, then it's obvious that the missile never actually needs to hit it's target and probably explodes using an area of effect (Source: it's a nuclear warhead), As Paul said if a target could actually dodge the explosion the current mechanic would make sense, I can't really find data for the explosion speed of Large nuclear weapons, but the Wikipedia page for project Orion mentions plasma velocities of 30,000 km/s, and since the energy of the explosion decreases as the fireball expands then I imagine you would need to be dead accurate in order to actually hit anything travelling at trans newtonian velocity. Furthermore this is pretty well represented by the fact that missile impacts punch distinct holes into armour which would require the detonation occuring close enough to the affected armour to send energy into the armour at a higher rate then that at which it can either absorb or reflect energy, going back to project orion it shows that under the right circumstances nuclear weapons won't destroy heavy armour as long as it's detonated far enough away from it.