Re: Relevance to Solar Wind
"The concern I have with solar wind is that I suspect (I'm not an astrophysicist) it is more variable and perhaps more significant."
It is more variably, but pretty much entirely insignificant. At 1 AU, which is the relevant distance from the Sun for things that might hit Earth, pressure from solar wind is around 1 nPa. So for a body with a Sun-facing area of 1 km^2, that's a 1e-3 N force, and a consequent acceleration of ~ 1e-14 m/s^2. In comparison, the Yarkovsky effect on the same body will cause a force of 0.25 N and acceleration around 1e-12 m/s^2. Far from being more significant, solar wind is on the order of a thousand times less significant. Yes, it's more variable, but any spikes are very short-term events. A CME might increase the solar wind by an order of magnitude or two, but that still only means it might be about the same strength as the Yarkovsky effect for a day or two, compared to being that strong all day, every day, for millions of years.
As for lacking information, we're really not. Things like CMEs happen on long enough timescales, and across large enough volumes of space, that we don't have any problem knowing they've happened even when the bulk of an event's activity occurs on the far side of the Sun from us.
"Ultimately, I would have thought that there is an intrinsic lack of detailed knowledge that would make orbital predictions (especially of medium-sized asteroids, big enough to be dangerous to Earth and also small enough to have larger orbital variations) markedly unreliable beyond a modest timeframe."
Absolutely. Orbital mechanics is a tricky business. The question comes down to what you consider to be a modest timeframe. Taking one of the examples from the article, asteroid 1950 DA has issues with the Yarkovsky effect because it creates some uncertainty in exactly where it will be 800 years from now. It would be utterly irrelevant to an object predicted to hit us within any living person's lifetime, but astronomers tend to consider things happening on a scale of centuries or millennia to be disturbingly fast. These kinds of tiny effects aren't at all relevant to the average person, it's just that scientists generally like to know things as accurately as possible even if it's something that will only affect distant descendents.