In support of your almost off-hand remark that "while all states may benefit when federal education policy-makers push good models, they will also all suffer when they push bad ones", I offer the California math standards as a case in point.  While our state Secretary of Education often touts them as "world-class", in my opinion (as a math educator who sees their impact in classrooms) they are terrible.  I'll spare you chapter and verse, but could cite instances where they are developmentally inappropriate (in their handling of place value and fractions and "algebra for all" at 8th grade), and other instances where they seem to avoid and trivialize mathematically important unifying concepts (e.g., proportion).  Tight conformity to these standards is contributing to the ills brought by NCLB.  I keep trying to find the good in them, and have managed to locate a couple of topics not handled too badly (e.g., integers), but overall have not revised my opinion of them when they were originally rolled out some ten or so years ago: they are terrible.  And I maintain that bad standards are worse than no standards.
Also: do you favor national standards, or a national curriculum?  Because national standards will not necessarily do much to alleviate the issues  around student mobility, which to me is the strongest argument for having standardization at either the state or national level.  And given the stranglehold of pacing guides and scripted curricula being imposed in California, and the pernicious effects of those--I would strongly argue against any national curriculum.  
There's irony in the fact that this whole standards movement was started by the 1988 (if I remember correctly) NCTM Standards.  The storm of controversy and reaction against them in the Math Wars (not as vicious as the Literacy Wars, but still vicious and destructive) resulted in the god-awful California standards we have today, through a not-so-pretty CA political process that I would hate to see exported and magnified to the national level.  
As with so much in the ed biz, you might get a majority of people agreeing that we need national standards, or a national curriculum--as long as those national standards or curriculum imposed their own values, biases, and beliefs.  But we have little agreement, even among educators let alone politicians or the general public, regarding those, even in math, let alone social studies or literature or science (evolution, anyone?).
I say we need more, not less, variety in public education.  Contra Arne Duncan, I do not think we know enough to be imposing one model, or even one set of curriculum standards (at least interpreted as CA's state standards are) on everyone.  Further, I do not believe that "one size fits all" in education, or ever will.  No one model will be good for everyone, regardless of its virtues.  I don't exactly know how to get there, but I'd like to see more, not less, control at the local level, allowing teachers, administrators, parents, and students trying out different things and going with their own strengths.  Not complete lassez faire (did I spell that right?  Seems to me there should be another "i" in there somewhere), but allowing people to make appropriate decisions for themselves, perhaps in collaboration with others in the system.  How to achieve this, in terms of a policy environment, is what I don't know.  Perhaps if we could agree on that as our common goal, we'd be able to figure out a way to do it.  But I'm not holding my breath.  Too many of us think we know what everyone ought to be doing (including me, all too often, I fear.)
 

In a not particularly good nutshell, standards set the goals, the curriculum defines how you get there.  Though of course, it's not that simple.  For instance, let's say we want all students to have understanding and facility with fractions by the end of eighth grade (and trust me, middle and high school math teachers would be thrilled if that happened), that would be a standard.  Defining that in second grade, we do "equal parts", in third grade we do unit fractions, in fourth grade we introduce non-unit fractions and equivalent fractions, in fifth grade we do operations with fractions, in sixth grade we do it all over again with mixed numbers--that would be somewhere between standards and a curriculum.  The full scope-and-sequence for each year (usually provided with the textbooks) would define a curriculum.  Unfortunately, the mobility problem is only addressed by the full curriculum.  Let's say every state has decided to include "equal parts" in the second grade--but in one state (or distirct) the topic is done in November, and in another it's done in March, a kid who moves from the latter to the former in January will miss it altogether.  Especially if teachers are rigidly tied to a pacing guide dictated from above, which is what's happening in CA these days.
For a much better discussion of standards vs. curriculum, read the preface to the old NCTM Curriculum Standards.  I can't remember if they kept it in the new Principles and Standards or not.