Recreating the H2 limiter on a PC is reasonably easy if you have suitable software. Personally I mostly use "Reaper" (
www.reaper.fm) for audio post production work. It currently costs a modest $40 but that's likely to increase shortly when version 2.0 is released - which will be a free upgrade from version 1.0
Once you've installed a program like that which can use VST effects, you can either use the effects that come with it or download others from the net - many of them free. For instance, the "Classic" range of effects (
http://www.kjaerhusaudio.com/classic-series.php) includes a very good and simple limiter, as well as a compressor and reverb, etc. Many other free limiters are available, and compressors, each with their particular characteristics. Apply limiting or compression after the recording is done does enable you to try the various options non-destructively - using the H2 limiter during recording means you're stuck with the result forever.
But it's often going to be better to normalise the recording - so that the loudest peaks just get to the top of the scale at 0dB - rather than limit. Limiting does reduce the dynamic range, and typically is used when you're not quite sure how loud the loudest peaks will be. Normalising can only be done after recording when you know the highest peak, so in postproduction that's the way to get the recording as loud as possible without compressing the original dynamic range between the quietest and loudest parts of the the performance.
As for the relative crudity of the H2 input level control - the three position switch - well, that's one step more than you get with a Hi-MD recorder, which simply offers high and low. I'm not sure about other such devices. Small inexpensive rotary input level controls can be hard to set accurately and may give rise to scratchiness in time.
I believe the switch on the H2 operates in 10dB steps. That means that the worst possible case would be a recording peaking at say -9.9dB (if you'd raised the sensitivity 10dB you'd have just clipped the recording by .1dB). That's no disaster, realistically - it means that the digital noise floor will end up 10dB higher than it would if you'd peaked to 0dB but the analog noise floor is going to be much more significant anyway. And in real-world conditions you'd probably actually occasionally peak a bit higher than -10dB.
I record a lot of classical music, professionally, and on my mic preamp gain control I have made a little red mark which represents the level that in almost all cases I use. So although I have a rotary control, if it had just one click position at that red mark, it would very often be all that would be required. That's because if I'm recording a symphony orchestra the main mic pair will be high up and back a bit. If it's a string quartet, the mics will be much closer. So the sound level arriving at the mic gets evened out by using the natural attenuation of distance. Performers ask me what they should play in rehearsal for me to get the levels right, but I tend to tell them that I know the level anyway! So, if the H2 designers have made a good choice of the preset levels available, it should be no disaster in practice.