If you choose to go the Zeo route, they still work and are useful. Which to choose (Alarm vs Mobile) depends on what you want to do with your gathered data.
- Using the Alarm version, you'll have to move your data "manually" via sdcard (or possibly a serial port) to a computer to use the data; the Alarm model has limited analytics on the clock itself. However, the Alarm version offers real-time data via the serial port, including access to the actual digitized waveform, which has allowed some to do some add-ons to the Alarm for lucid dreaming applications.
- Using the Mobile version, you'll have access to your data on the Android device with graphical results, have some analytics right on the Android device too if you pick up the ZeoCompanion app, and can easily send your data to a computer via email for more extensive analytics. But to conserve battery while using bluetooth to transmit to the Android device, the Zeo Mobile headband sends sleep data only every 5 minutes, making it less useful for applications like lucid dreaming. And the Zeo Mobile headband cannot send the raw waveform nor accelerometer data.
So if you intend to analyze your data and maybe perform some sleep experiments with supplements and other sleep aids, the Mobile version will make that much easier. If all you want is a sleep score each night, then either version will work for something that simple.
The Zeo's headband can be repaired/replaced, and from your question I think you've found the How-To's on doing that. The headband pad's life can also be extended simply by being diligent in cleaning the pads and your forehead each night.
As Al_Luckow mentioned, most people stop using their Zeo after a month or so, simply because just accurately knowing how much deep vs REM vs Light you got each night is really quite useless. Having that data is only really useful if you are actively trying to improve your sleep, and are trying out different sleep aids, techniques, and supplements, and want to see quantitatively how they improve your sleep rather than a subjective "I sleep better last night ... I think" kind of result score.
Most of the current crop of users of the Zeo Mobile (2,600 to be exact) were actively involved in sleep improvement supplementation experiments spawned by Steve Gibson and his "Security Now" podcast ... health and sleep improvement are side-projects of his that get occasional mentions on the podcast. But he's got a whole forum devoted to his Healthy Sleep Formula supplementation recommendation. They were all using the Zeo to test out specific supplements (and dosages of those supplements), recording their resultant quantitative effects especially on total Deep sleep and sustained sleep (quantity of over-night wake-ups).
Again as Al_Luckow mentioned and perhaps Luke_Sleepwalker implied, using an EEG tool may not really be necessary to meet your goals (which you did not provide us). And it may be a useful exercise to continue using other less accurate set-and-forget sleep assessment tools (like a smartphone or a sensor pad or a wrist-tracker) at the same time as using the Zeo. Then calculate a correlation between the Zeo's data and the other tool's data. You can then in the long-term empirically infer your resultant sleep still using these set-and-forget tools long after you've gotten bored with putting the Zeo on your head every night