As my disclaimer, I am one of those that reminds new (and not so new) applicants to the NIH for research funding to talk to the Program Officer. Early and often, just like voting. In the pre-submission phase you want to identify the line POs down in the Division and Branch structure of one of the funding Institutes or Centers of NIH who might be interested in your work. After review, the PO who was actually assigned to your application can give you invaluable feedback about how the review of your application went down. So I recommend calling this person.
Take their comments, however, in context. They don't know everything. Even if they are sitting in the room, paying attention to the discussion of grants, this does not always mean they truly understand what is going down.
One of the times I get really frustrated with POs is when they are so fixated on the objective truth of peer review. They often act as though they believe that the review process really does work nearly perfectly. Most often when it comes to newbies. Consequently if you are coming up short on your proposals, in their worldview they think you are "not writing well enough". Or need to (somehow without funding) provide more/better preliminary data.
So this advice gets reflected back on the poor applicant who 1) drives herself crazy trying to "improve" her writing (which is just fine already) or 2) goes back to the lab to find that perfect figure which which will guarantee this grant gets funded (no such thing).
Disastrously, this prevents said newbie from doing what she really needs to be doing which is to submit multiple good-enough grants. In the face of budgets which allow the funding of only a subset (a third? quarter?) of the grants which are excellent and interesting and impactful and all that jazz, review becomes variable. Meaning the difference between making it into a fundable score and just missing a fundable score takes on the appearance of chance. The only way to beat such odds is to give yourself more chances at the game. This means writing and submitting multiple applications (on different topics, of course).
Don't mistake me. There IS a learning process for grant writing and it remains good advice for the new (and not so new) investigator to seek feedback from peers prior to submitting a grant and when doing the post-mortem after receiving the summary statement. But this can't be taken too far. At some point, you are just driving yourself crazy with conflicting (good) advice from people who are in different situations from yourself or each other. And anyway, you can apply the stylistic and structural advice you have received to new application just as well as to the revision of your existing application, right?