I would say they are able to do so as a result of inquisitor-specific abilities that are insulating them from insanity, or that these characters are actually high-functioning 100 insanity characters. A character at that level is 'unplayable' in the FFG tabletop, but that doesn't mean that a character necessarily breaks into a gibbering mess at this point. The Black Crusade book actually suggests that renegades are those who have gone beyond the insanity and corruption scales from other games and come out on the other side with newer, more grievous metrics.

As one item, I feel that calling for an Exterminatus should always increment insanity.

They talk about them in Beyond.

I'm making the distinction assuming that shooting is your intent. A target that you can visibly see, who is an active threat, and does not have cover. It should never ever be optimal to use overwatch on a target like that - an enemy who is actively and openly engaged in combat against you in an open field.

Notably, an Inquisitor should have very high Willpower and has access to additional talents related to resisting the progress of insanity.

The real question here ought to be - does using a sword vs. an axe make you better at parrying with your shield? Does using a shield make you better at parrying with your sword? Alternatively, is parrying weapon-agnostic? If a parry test doesn't care about which weapon you use, how do you resolve Power Fields? If I wield one balanced and one unbalanced weapon, do the benefits and penalties cancel, or can I parry with one and disregard the other? These questions should elicit the outline for how each weapon combination statistically behaves.

As far as the RaI on Overwatch, I feel it is against the spirit of the game any time Overwatch is used on targets in open ground. If a target is in open ground, it feels intuitive to either just shoot, or suppress. It makes sense to Overwatch when another party member is performing a task, or when an enemy is in cover. If you take up an understanding of the rules that violates this intent, you would be better served to house rule the game back on track.

In DH2, things like "armor plates" can begin to prompt issues with subtlety. Ultimately though, soak is very strong and needs to be strictly controlled. Mutual exclusivity or non-stacking also inhibits munchkin-ing, which can quickly get out of control in this kind of crunchy system. One bit that we have enjoyed about bionics is allowing "sub-systems" for higher grade bionics that allow the incorporation of additional wargear if you can requisition a high quality version of that gear. Best Craftsmanship Limbs with Best Craftsmanship Recoil Gloves = Recoil Dampening Arms.

I also ascribe to this. Allowing the two to stack here gets ridiculous quick, I tend to think that DH2 lacks "The Flesh Is Weak" for good reason.

I haven't seen any errata newer that July 2015... where is a newer one printed?

In our games, we found that this was an issue and ruled that critical damage is only removed by Extended Care. If you aren't critically wounded, first aid can probably fix all of your issues, but crits stick. We do reset the critical tracker on fate burn.

Be mindful that the final bonus/penalty cannot be greater than +/-60. Other than that, accurate is just pretty good. It kills singular entities and loses to large groups. An accurate weapon is still less effective per rarity against armor than a melta weapon or a krak missile. A flamer is still better at killing a crowd of twenty knife-wielding cultists. Accurate weapons are VERY powerful if you never catch players off-guard. Surprising the players and upsetting their plans is important sometimes.

Forgetting the book on this one, you intuitively only gain benefits for the weapon you are using to parry. Defensive should be a quality that overwrites balanced, in the same way that unwieldy overwrites unbalanced. The books only go so far, this is a good occasion to common-sense house rule.

If a vehicle's operator cannot take actions, I would put it out of control, determining it's direction accordingly (speed tactical +1d10). I would determine randomly if the operator lead-footed or slipped off the pedal. If leadfoot, rerolling each turn until eventually the vehicle is no longer getting acceleration. After that, cut the speed by about 1/3rd per round. Dragging the operator out of the seat may take time depending on the nature of the vehicle, but grabbing the wheel from him should be a quicker option for some course correction.

Most roles don't get to change their core gameplay with their special ability. Many of them are about consistency or throughput in a dedicated sphere. Most role benefits are something you graduate from. As high rates of success become more standard for your character, the benefits of auto-succeeding or setting a certain number of successes diminish. Notably, Move and Shoot is very powerful in early play, where running away may be more necessary due to minimal equipment, and enemies may lack powerful long-range armaments. Later on, it's a tool in the kit, but it shouldn't be your primary mode.