I've been following Newlife for a few years and poked around with a decompiler at one point, comparing the differences in code between a couple of versions. My take is that SO is not lying when they say they put hours into it, it's just that Newlife has deep-rooted design issues that make it extremely cumbersome to develop new scenes for.
A lot of Newlife's writing effort looks to be wasted on the many combinations of character traits and NPC traits, none of which can be seen without multiple playthroughs. The inflexibility of character traits is an unusual design decision. A lot of games make gradual changes to the main character's sexuality a staple of their gameplay. If Newlife was refactored / reworked to a more flexible system of tracking behaviour characteristics, one that would allow for unlocking content based on character choices instead of just decisions made at the start, I think players would get a lot more mileage out of the variations and their character would seem less static.
Unfortunately, reworking Newlife substantially would be difficult, because so much of the text content looks to be tied up in blocks of code. There was an attempt to create a higher level scripting language to allow third parties to create custom scenes, but instead of developing that idea further and adopting it wholly, the majority of SO's content is formed by concatenating java strings. Trying to connecting the variables tracked by the game to all the variations of text in a way that makes grammatical sense is an extremely high workload task without using some sort of higher level language.
TLDR: I can believe SO when they say they work hard on Newlife, but I think the resulting progress is massively underwhelming because of flawed game design decisions that are difficult to reverse.