When recording a phone interview, you almost always have to set the recording level for the loudest of the two voices, and then use a dynamics compressor/limiter to get the perceived volume of the two voices to match. I believe there is a dynamics compressor/limiter in your H4N — it’s been a while since I owned one.
Also ask them to speak directly into their phone, and ask them to try to speak as though they were speaking to you in the room they are in. Sometimes people tend to speak in an intimate, right-next-to-your-ear fashion on the phone that is hard to record.
One tip is to listen through the recorder. Don’t listen to the landline handset. Then you can move away from the microphone a bit (as long as the person on the other end can hear you) and increase the recording gain to capture more of the other person, and then engage the dynamics compressor/limiter which will bring the volume of the lowest voice up to where the highest voice is. This also has the benefit of enabling you to monitor what the recorder is actually hearing as you conduct the interview.