数据分析方法求解

Call:
glm(formula = flag ~ X1wt + X1zf + X2zf, family = binomial(),
    data = datasample)

Deviance Residuals:
    Min       1Q   Median       3Q      Max  
-1.8524  -1.2679   0.7981   0.8894   1.4281  

Coefficients:
              Estimate Std. Error z value Pr(>|z|)   
(Intercept) -137.59437   34.97788  -3.934 8.36e-05 ***
X1wt           0.90419    0.28296   3.195   0.0014 **
X1zf           0.51834    0.32318   1.604   0.1087   
X2zf           0.04156    0.29319   0.142   0.8873   
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 812.58  on 632  degrees of freedom
Residual deviance: 789.01  on 629  degrees of freedom
AIC: 797.01

Number of Fisher Scoring iterations: 4


以上是我用logistic回归预测机器表现指标是否在某一范围内，从P值来看好像除了X1wt这个指标对方程贡献显著外，其余两个（X1zf 、X2zf  ）指标都不显著，这样是说明模型拟合效果不好吗？那接下来该怎么办呢？