Electronic timers are never lighter than mechanical timers by the time you add the battery and electromechanical bits to the electronics and make the lot waterproof and crashproof. The electromechanical drive, a servo or solenoid, must convert the signal from the electronics into a control movement and needs to work reliably against surprisingly high forces. This in turn means that the battery needs a decent capacity if it is to last more than one or two flights. Finally, you need some means of setting the timer and starting it at launch. The various bits and associated wiring take up more space in the fuselage than mechanical systems.
All the current timers are multi-function microprocessor based designs using RC servos as output devices. Its just not practical or cost-effective to build a multi-function timer in any other way. They use a Palm Pilot or other PDA to set times and servo output movements because this eliminates set-up switches, saving weight and adding reliability. Miniature RCHLG servos (FMA SS-80, etc.) are used as output devices. The indoor RC servos, while even lighter and smaller, do not have the power needed to drive systems on an outdoor FF model against the spring tensions needed to assure positive trim settings.
Electronic timers are not necessarily more reliable than mechanical timers. For a start, if your release mechanisms are badly made and set up (levers that stick, lines not free running, hooks near the timer for the line to catch on, etc., etc.) an electronic timer will not solve any problems. The timer itself is not foolproof either. While they don't break springs or get grit in the cogs they do have their own unique problems: faulty batteries, fractured wiring, high resistance contacts.
I've lost three models with electronic timers, all flyaways, and had three almost losses that happened to come down. One loss was almost certainly a battery failure and the last two were probably a bad microswitch. The other incidents were recovered and the failure reasons analysed. I haven't had any repeats of those problems showing that I do sometimes learn from my mistakes. However, that's only six incidents in 20 years. Not so bad a record.
A straight 555 based timer is often suggested by electronics engineers. Sorry, we been there, done that and this approach has been proved impractical for several reasons. These include the difficulty of getting accurate d/t settings with a rotary pot (non-linear analogue settings, temperature drift) and the problem of battery size; unless you add more electronics (at least a 556) you'll be stuck with a permanently on solenoid and resulting high battery drain.
I've known more or less exactly what Thomas Koster has been doing with electronic timers between 1977 and 2000 and have a fair bit of experience using them myself. Over this time I've discovered (or been told about) most of the pitfalls and blind alleys in this area. The simple fact is that developing a lightweight, small and reliable timer is very difficult and often counterintuitive. For instance the reason Thomas originally used an external control box was because he was unable to find an on-model setting system that could survive the vibration of a running F1C for more than one or two flights. Similarly, he found that naked CMOS circuitry will not operate reliably near a glow motor, hence the necessity to pot the electronics. Methanol vapour in the air lowers its conductivity to the point that CMOS chips self-trigger due to conductance between adjacent pins.