Now is the time to change something. He's gotta rewrite the whole game anyway, why not rewrite it better?
The risk is that the complexity of porting the existing rule set to C# combined with the complexity of changing the rule set leads to a more complex overall task. Trying to combine the two goes against most modern philosophies of good software engineering practices. I personally think this has been demonstrated with the ground combat stuff: Steve was chugging along through the various functionality sectors until he got to ground combat, at which point my perception is that the progress got bogged down. Some data: the original "I am seriously considering removing PDCs" post was on Sept 17, 2017 (on page 70 out of 93 in this thread), so he's been on ground combat for four calendar months and 25% of the posts in this thread, which is probably more time than it would have taken if he'd simply transcribed PDCs and ground combat.
A more telling example of the "change while transcribing" failure mode: the Pulsar 4x project seems to have fallen prey to this it. My recollection/perception is that they started out wanting to do a straight port of Aurora to C#, but figured "why not improve the game mechanics along the way". They seemed to have bogged down about halfway through; I haven't seem much activity from them at all for the last year or two.
That being said:
1) Steve is good at this stuff (writing Aurora); he's been doing it for many years.
2) Ground combat IS an isolated system, so he can code it up and get out of it. In general, Steve seems to be doing a really good job of doing minor and isolated tweaks to systems as he codes them up (e.g. the refueling changes), so that he's doing "transcription with cleanup" as he goes, rather than "write a whole new game". In other words, I think he's mostly been striking the right balance.
3) Steve's already written one game (VB6 Aurora) and gotten it to completion, which demonstrates the tenacity and ability to complete the job, so there's a good chance he'll complete C# Aurora too.
4) (Most important) It's Steve's free time, so if he wants to spend it working out ground combat mechanics before getting a playable version of Aurora out then that's his prerogative.
So I suspect the ground combat stuff has probably slipped the schedule by a month or two, but if that's what Steve wants to do that's his choice. Hopefully this is the only sector for which he's planning a major rewrite, and he can get back to straight transcription to get a working version out. If not though, c'est la vie.
John